Instructions to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
# Run inference directly in the terminal:
llama cli -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
# Run inference directly in the terminal:
llama cli -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
# Run inference directly in the terminal:
./llama-cli -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
# Run inference directly in the terminal:
./build/bin/llama-cli -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Use Docker

docker model run hf.co/apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

LM Studio
Jan

vLLM

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Ollama
How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Ollama:
```
ollama run hf.co/apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
```

Unsloth Studio

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 to start chatting

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Docker Model Runner:
```
docker model run hf.co/apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128
```

Lemonade

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Run and chat with the model

lemonade run user.DeepSeek-V4-Flash-0731-DS4-Quality128-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128 with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default apetersson/DeepSeek-V4-Flash-0731-DS4-Quality128

Run Hermes

hermes

Atomic Chat

apetersson commited on 4 days ago

Commit

a40bb77

verified ·

1 Parent(s): 78623a7

Add files using upload-large-folder tool

Browse files

Files changed (7) hide show

.gitattributes +2 -0
BUILD_MANIFEST.json +1052 -0
DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf +3 -0
DeepSeek-V4-Flash-0731-DS4-Quality128.gguf +3 -0
PROVENANCE.md +19 -0
README.md +379 -0
SHA256SUMS +5 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf filter=lfs diff=lfs merge=lfs -text
+DeepSeek-V4-Flash-0731-DS4-Quality128.gguf filter=lfs diff=lfs merge=lfs -text

BUILD_MANIFEST.json ADDED Viewed

	@@ -0,0 +1,1052 @@

+{
+  "schema": "deepseek-model-tools.build128.v1",
+  "artifact": "ds4-quality128-original",
+  "output": "DeepSeek-V4-Flash-0731-DS4-Quality128",
+  "engine": "ds4",
+  "dspark": true,
+  "created_at": "2026-08-01T16:00:13.755362+00:00",
+  "source": {
+    "path": "/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731",
+    "only_weight_source": true,
+    "source_kind": "original-huggingface-checkpoint",
+    "config_sha256": "6c8f3d2d3b48707541b88f32f22ef3f0f8a6b57d8523281e2b8d3cdb0ae9a023",
+    "index_sha256": "98efab455cf08dfbbbaaba6f570e1bf10bf927d2b4c3c453a59c2f6f0e3be92b",
+    "upstream_model": "deepseek-ai/DeepSeek-V4-Flash-0731",
+    "upstream_revision": "9e165c30e2704aec5d9d593cce3eebd58bbef1cb",
+    "identity_cache": "/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731/.cache/huggingface/download",
+    "weight_shards": [
+      {
+        "filename": "model-00001-of-00048.safetensors",
+        "sha256": "f3668ba4cccf1ca6a7eb84e888fb92c1cdc7204d472ba9db771e6fd3abf6b874",
+        "upload_mode": "lfs",
+        "bytes": 1059061856
+      },
+      {
+        "filename": "model-00002-of-00048.safetensors",
+        "sha256": "77b26c939a0e25b3113c8d6bb04e1901a748bd4a7d2589e3bfdaabdf1e9bba14",
+        "upload_mode": "lfs",
+        "bytes": 3566321192
+      },
+      {
+        "filename": "model-00003-of-00048.safetensors",
+        "sha256": "412abf4c906faadc221ef0cb50f90fe20bde8454a08ad4dc2364b6b79e7fda5c",
+        "upload_mode": "lfs",
+        "bytes": 3566321192
+      },
+      {
+        "filename": "model-00004-of-00048.safetensors",
+        "sha256": "9610f56bc587fb0ff9a8b68a60299482ee8c433fe5b5587e4257aca98add4a2e",
+        "upload_mode": "lfs",
+        "bytes": 3596229272
+      },
+      {
+        "filename": "model-00005-of-00048.safetensors",
+        "sha256": "f87a5ac7b8becc31f9c3169afd3a6f33fb82b4af9e21022e3755a10bc28f0180",
+        "upload_mode": "lfs",
+        "bytes": 3568768976
+      },
+      {
+        "filename": "model-00006-of-00048.safetensors",
+        "sha256": "4a4f3764e3fc772b9fba67f0a44ef68e18f178b6f00faa80b75db549e51894cd",
+        "upload_mode": "lfs",
+        "bytes": 3590024776
+      },
+      {
+        "filename": "model-00007-of-00048.safetensors",
+        "sha256": "df81bb80e27a689e01fa579eebd6499f86e0b6105f7fea18961aa5eebbbee9bc",
+        "upload_mode": "lfs",
+        "bytes": 3568768976
+      },
+      {
+        "filename": "model-00008-of-00048.safetensors",
+        "sha256": "224968d2b27f8669365ec08657a768dfec40da0585f85f302a31495931f6a526",
+        "upload_mode": "lfs",
+        "bytes": 3590024776
+      },
+      {
+        "filename": "model-00009-of-00048.safetensors",
+        "sha256": "04d69ef1071fff8721c62968c200a5583122b59b015e9ef9b2978bfed271b2b7",
+        "upload_mode": "lfs",
+        "bytes": 3568768976
+      },
+      {
+        "filename": "model-00010-of-00048.safetensors",
+        "sha256": "627145f4ebeb1cc3f5bdd03416b8cb7370b3c96974853cbeb8e5516ad5713e49",
+        "upload_mode": "lfs",
+        "bytes": 3590024776
+      },
+      {
+        "filename": "model-00011-of-00048.safetensors",
+        "sha256": "e4b8e601dcbebe902e0102e7b098b670a121cb8b9564dd719fc41d782c8416e0",
+        "upload_mode": "lfs",
+        "bytes": 3568768976
+      },
+      {
+        "filename": "model-00012-of-00048.safetensors",
+        "sha256": "64ed4e5f6126ba029c462c9d5fca0fc907c5f855b4ba01194d79560f6db16e42",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00013-of-00048.safetensors",
+        "sha256": "8dfe199d07c07ddd141c2c0136a2237f1161250a1a03ebe8deaabac93440da1d",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00014-of-00048.safetensors",
+        "sha256": "45db2f540f825f92453c50335e49aede58cca56bc578d1787c12a0fbca6593e5",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00015-of-00048.safetensors",
+        "sha256": "5810381a0f05b7381c002d299ed6ac19e42eba8070dd17e2703546944d84f292",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00016-of-00048.safetensors",
+        "sha256": "e0530b7024771b0ce2df9b40bcc2232578f3300178487ec216863b0b2835617b",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00017-of-00048.safetensors",
+        "sha256": "ed11130247118b185ade893c0109bad896dd394cb1e066ce4fce044176261d94",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00018-of-00048.safetensors",
+        "sha256": "e393fea96da2a3414ef089354fc32e1c8891954de40958c84d2c2ecf80365b25",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00019-of-00048.safetensors",
+        "sha256": "a74ca4d3e8e82ce20c458bb8b1900110b753793ad4c58d08f38995f719c616f7",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00020-of-00048.safetensors",
+        "sha256": "9f556769926e60309e8defe45ab59fc8b26ae460d30c190cd746a3d78c11e2c2",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00021-of-00048.safetensors",
+        "sha256": "1671cce7f90d781f796b5ca6bf32dd1aeb740abcb2735e41ffd28f62485ce005",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00022-of-00048.safetensors",
+        "sha256": "decd67a4bd97a75fa36861d2ad3067afeefa6a04a20da997fe6c19f171e70132",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00023-of-00048.safetensors",
+        "sha256": "c61a3e179cdbee19bfb8cbc4e111928ce2f1e1f0f4729d7c0cd5634354a4689d",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00024-of-00048.safetensors",
+        "sha256": "fc27aeb4233534f6f7781dcfe57127a3908ae10fc025c5d86dc0682057f8b2fe",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00025-of-00048.safetensors",
+        "sha256": "a66b6b8d5821b68f5b511e4f91e12025cd07d0fa6d0b71e722d825a2d6d878ca",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00026-of-00048.safetensors",
+        "sha256": "657b89314fbaf6eee4acce24b3baf7e5fd2c5986a96ad85b08d90539cde869fe",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00027-of-00048.safetensors",
+        "sha256": "fb01f21a0da0446b0bdf25a127ab19a6b06006acd8735f06a9ebfe34423fd7f5",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00028-of-00048.safetensors",
+        "sha256": "b2fd5cbbb639f16e673bc484e5cca16b52a58bf2ab4bd62592e0c5408712ad7c",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00029-of-00048.safetensors",
+        "sha256": "9ec2fdf900275daeac0980490c5c731cc7868b151ce1de5698f48418de4fa5f0",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00030-of-00048.safetensors",
+        "sha256": "9ed3c317bf967d32133ad3a068ee4c56aae9784bc8b7da694482437f37dc1782",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00031-of-00048.safetensors",
+        "sha256": "d5078c3fca3e6370043606ead7856e0b8fe67a9aab52c415769f29934c4d7f5d",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00032-of-00048.safetensors",
+        "sha256": "163653848f002718d3deaa6ce48885483fc1f2e12e50e44a47477c73ccd91393",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00033-of-00048.safetensors",
+        "sha256": "f2cffd43f2a5f491f4691f8694e6bc08239158e143ae7063dc04f0eb0259214a",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00034-of-00048.safetensors",
+        "sha256": "0f94945121474cfdb6a9ab175914d3811ffbf08e6cc54082e14e473c755d18d8",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00035-of-00048.safetensors",
+        "sha256": "9cb6a316989f7c7385e3ec2bd42ffe766ee126c70ef3466742849982ee1b0f0f",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00036-of-00048.safetensors",
+        "sha256": "7e6761421fe944c2143eb897b983085891b421c148c6c17fb5cd8eaa9bdaa497",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00037-of-00048.safetensors",
+        "sha256": "a59d662f1143596d56c452a1230b717ce43edf678207398c573a2503b0f72c91",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00038-of-00048.safetensors",
+        "sha256": "137fa617a74ba8e73fd76bb1010c7a85d791aaa150006cb66faa04a83e9e730f",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00039-of-00048.safetensors",
+        "sha256": "a29af1aa519d7ce726235ea2c2b38146d756290cda7a82d90c4d4438155b53e4",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00040-of-00048.safetensors",
+        "sha256": "8bc93d8a7d1987dc86b14e22b1d8f42ec31da92c56edd9f312daf43f33a6a206",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00041-of-00048.safetensors",
+        "sha256": "fd312e7fdd6cb5796df356a7f0314f124851dc149991e9fb02c5bed45cc4ba05",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00042-of-00048.safetensors",
+        "sha256": "4d19bf368083c9a183cb0849f316ec17b62f859ca824c0586e779657efb6e6a6",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00043-of-00048.safetensors",
+        "sha256": "b7103842ceb70848f9804f55c193d6a57f43174a587cba42b61b5c1bc4e1303d",
+        "upload_mode": "lfs",
+        "bytes": 3568770544
+      },
+      {
+        "filename": "model-00044-of-00048.safetensors",
+        "sha256": "422d3889fa20c238b7f97464c14df0bcf3328f189c294f41a3a334421dc560c7",
+        "upload_mode": "lfs",
+        "bytes": 3590026352
+      },
+      {
+        "filename": "model-00045-of-00048.safetensors",
+        "sha256": "a5be6aed7b84fc87ec42b5d24ba0b0d67f253a3906fcd99c13f4f7be5958fc00",
+        "upload_mode": "lfs",
+        "bytes": 1059332516
+      },
+      {
+        "filename": "model-00046-of-00048.safetensors",
+        "sha256": "5db924ca907e0d93acd975bd5079c3662717f9ac709f23d079bd8f816d29d9dd",
+        "upload_mode": "lfs",
+        "bytes": 3610455184
+      },
+      {
+        "filename": "model-00047-of-00048.safetensors",
+        "sha256": "62816173f9f6e136b20b48e3b6f16613ac9ea02b5603f636928b253244a548bd",
+        "upload_mode": "lfs",
+        "bytes": 3560111960
+      },
+      {
+        "filename": "model-00048-of-00048.safetensors",
+        "sha256": "cc43742bd24ae6bcdea343a91442f6f66aed2cfebcc6b235470204851ce2f8a9",
+        "upload_mode": "lfs",
+        "bytes": 3692775244
+      }
+    ],
+    "weight_shard_bytes": 166886535336,
+    "weight_shard_hashes_verified": true,
+    "weight_shard_hashes_verified_at": "2026-08-01T14:21:34.126450+00:00",
+    "post_hash_stat_snapshot": {
+      "model-00001-of-00048.safetensors": {
+        "bytes": 1059061856,
+        "mtime_ns": 1785503207610000000,
+        "ctime_ns": 1785503207610000000,
+        "device": 16777242,
+        "inode": 1227200
+      },
+      "model-00002-of-00048.safetensors": {
+        "bytes": 3566321192,
+        "mtime_ns": 1785503526950000000,
+        "ctime_ns": 1785503526950000000,
+        "device": 16777242,
+        "inode": 1220339
+      },
+      "model-00003-of-00048.safetensors": {
+        "bytes": 3566321192,
+        "mtime_ns": 1785503449850000000,
+        "ctime_ns": 1785503449850000000,
+        "device": 16777242,
+        "inode": 1220314
+      },
+      "model-00004-of-00048.safetensors": {
+        "bytes": 3596229272,
+        "mtime_ns": 1785503514180000000,
+        "ctime_ns": 1785503514180000000,
+        "device": 16777242,
+        "inode": 1220289
+      },
+      "model-00005-of-00048.safetensors": {
+        "bytes": 3568768976,
+        "mtime_ns": 1785503576410000000,
+        "ctime_ns": 1785503576410000000,
+        "device": 16777242,
+        "inode": 1220288
+      },
+      "model-00006-of-00048.safetensors": {
+        "bytes": 3590024776,
+        "mtime_ns": 1785503626970000000,
+        "ctime_ns": 1785503626970000000,
+        "device": 16777242,
+        "inode": 1220364
+      },
+      "model-00007-of-00048.safetensors": {
+        "bytes": 3568768976,
+        "mtime_ns": 1785503569660000000,
+        "ctime_ns": 1785503569660000000,
+        "device": 16777242,
+        "inode": 1220287
+      },
+      "model-00008-of-00048.safetensors": {
+        "bytes": 3590024776,
+        "mtime_ns": 1785503573510000000,
+        "ctime_ns": 1785503573510000000,
+        "device": 16777242,
+        "inode": 1220286
+      },
+      "model-00009-of-00048.safetensors": {
+        "bytes": 3568768976,
+        "mtime_ns": 1785503819100000000,
+        "ctime_ns": 1785503819100000000,
+        "device": 16777242,
+        "inode": 1257677
+      },
+      "model-00010-of-00048.safetensors": {
+        "bytes": 3590024776,
+        "mtime_ns": 1785503875280000000,
+        "ctime_ns": 1785503875280000000,
+        "device": 16777242,
+        "inode": 1412950
+      },
+      "model-00011-of-00048.safetensors": {
+        "bytes": 3568768976,
+        "mtime_ns": 1785503964350000000,
+        "ctime_ns": 1785503964350000000,
+        "device": 16777242,
+        "inode": 1439498
+      },
+      "model-00012-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785503994640000000,
+        "ctime_ns": 1785503994640000000,
+        "device": 16777242,
+        "inode": 1479217
+      },
+      "model-00013-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504255840000000,
+        "ctime_ns": 1785504255840000000,
+        "device": 16777242,
+        "inode": 2913063
+      },
+      "model-00014-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504220800000000,
+        "ctime_ns": 1785504220800000000,
+        "device": 16777242,
+        "inode": 2943079
+      },
+      "model-00015-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504267840000000,
+        "ctime_ns": 1785504267840000000,
+        "device": 16777242,
+        "inode": 2943120
+      },
+      "model-00016-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504275460000000,
+        "ctime_ns": 1785504275460000000,
+        "device": 16777242,
+        "inode": 3008574
+      },
+      "model-00017-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504389580000000,
+        "ctime_ns": 1785504389580000000,
+        "device": 16777242,
+        "inode": 3111729
+      },
+      "model-00018-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504430230000000,
+        "ctime_ns": 1785504430230000000,
+        "device": 16777242,
+        "inode": 3142001
+      },
+      "model-00019-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504408150000000,
+        "ctime_ns": 1785504408150000000,
+        "device": 16777242,
+        "inode": 3175731
+      },
+      "model-00020-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504589260000000,
+        "ctime_ns": 1785504589260000000,
+        "device": 16777242,
+        "inode": 3179314
+      },
+      "model-00021-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504739580000000,
+        "ctime_ns": 1785504739580000000,
+        "device": 16777242,
+        "inode": 3250866
+      },
+      "model-00022-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504807160000000,
+        "ctime_ns": 1785504807160000000,
+        "device": 16777242,
+        "inode": 3273972
+      },
+      "model-00023-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785504764960000000,
+        "ctime_ns": 1785504764960000000,
+        "device": 16777242,
+        "inode": 3280897
+      },
+      "model-00024-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785504875800000000,
+        "ctime_ns": 1785504875800000000,
+        "device": 16777242,
+        "inode": 3280896
+      },
+      "model-00025-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505025920000000,
+        "ctime_ns": 1785505025920000000,
+        "device": 16777242,
+        "inode": 3413097
+      },
+      "model-00026-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505045180000000,
+        "ctime_ns": 1785505045180000000,
+        "device": 16777242,
+        "inode": 3413096
+      },
+      "model-00027-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505045830000000,
+        "ctime_ns": 1785505045830000000,
+        "device": 16777242,
+        "inode": 3476335
+      },
+      "model-00028-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505098680000000,
+        "ctime_ns": 1785505098680000000,
+        "device": 16777242,
+        "inode": 3862697
+      },
+      "model-00029-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505337860000000,
+        "ctime_ns": 1785505337860000000,
+        "device": 16777242,
+        "inode": 3962848
+      },
+      "model-00030-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505329360000000,
+        "ctime_ns": 1785505329360000000,
+        "device": 16777242,
+        "inode": 3969579
+      },
+      "model-00031-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505278610000000,
+        "ctime_ns": 1785505278610000000,
+        "device": 16777242,
+        "inode": 6552978
+      },
+      "model-00032-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505422450000000,
+        "ctime_ns": 1785505422450000000,
+        "device": 16777242,
+        "inode": 6577364
+      },
+      "model-00033-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505518610000000,
+        "ctime_ns": 1785505518610000000,
+        "device": 16777242,
+        "inode": 9337694
+      },
+      "model-00034-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505646240000000,
+        "ctime_ns": 1785505646240000000,
+        "device": 16777242,
+        "inode": 9337693
+      },
+      "model-00035-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505632340000000,
+        "ctime_ns": 1785505632340000000,
+        "device": 16777242,
+        "inode": 9341904
+      },
+      "model-00036-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505677330000000,
+        "ctime_ns": 1785505677330000000,
+        "device": 16777242,
+        "inode": 9347816
+      },
+      "model-00037-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505900360000000,
+        "ctime_ns": 1785505900360000000,
+        "device": 16777242,
+        "inode": 9384464
+      },
+      "model-00038-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505965980000000,
+        "ctime_ns": 1785505965980000000,
+        "device": 16777242,
+        "inode": 9391578
+      },
+      "model-00039-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505970320000000,
+        "ctime_ns": 1785505970320000000,
+        "device": 16777242,
+        "inode": 9398495
+      },
+      "model-00040-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785505978650000000,
+        "ctime_ns": 1785505978650000000,
+        "device": 16777242,
+        "inode": 9411167
+      },
+      "model-00041-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785505946150000000,
+        "ctime_ns": 1785505946150000000,
+        "device": 16777242,
+        "inode": 10377561
+      },
+      "model-00042-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785506161290000000,
+        "ctime_ns": 1785506161290000000,
+        "device": 16777242,
+        "inode": 10394652
+      },
+      "model-00043-of-00048.safetensors": {
+        "bytes": 3568770544,
+        "mtime_ns": 1785506159880000000,
+        "ctime_ns": 1785506159880000000,
+        "device": 16777242,
+        "inode": 10401334
+      },
+      "model-00044-of-00048.safetensors": {
+        "bytes": 3590026352,
+        "mtime_ns": 1785506177170000000,
+        "ctime_ns": 1785506177170000000,
+        "device": 16777242,
+        "inode": 10405174
+      },
+      "model-00045-of-00048.safetensors": {
+        "bytes": 1059332516,
+        "mtime_ns": 1785506038960000000,
+        "ctime_ns": 1785506038960000000,
+        "device": 16777242,
+        "inode": 10451859
+      },
+      "model-00046-of-00048.safetensors": {
+        "bytes": 3610455184,
+        "mtime_ns": 1785506331530000000,
+        "ctime_ns": 1785506331530000000,
+        "device": 16777242,
+        "inode": 10459297
+      },
+      "model-00047-of-00048.safetensors": {
+        "bytes": 3560111960,
+        "mtime_ns": 1785506331790000000,
+        "ctime_ns": 1785506331790000000,
+        "device": 16777242,
+        "inode": 10459296
+      },
+      "model-00048-of-00048.safetensors": {
+        "bytes": 3692775244,
+        "mtime_ns": 1785506331020000000,
+        "ctime_ns": 1785506331020000000,
+        "device": 16777242,
+        "inode": 10464304
+      }
+    }
+  },
+  "profile": {
+    "path": "/Users/andreas/code/deepseek-model-tools/configs/dsv4-ds4-quality128-original.json",
+    "sha256": "bc7f6470a0cb6208addf2ac52806d976108517119b411023da3fd91b492b218c",
+    "content": {
+      "name": "dsv4-ds4-quality128-original",
+      "engine": "ds4",
+      "description": "Quality-first 128 GB DS4 profile for the official original DeepSeek V4 Flash 0731 checkpoint, preserving exact native MXFP4 routed experts on ten sensitivity-selected layers, with IQ2_XXS/Q2_K elsewhere and native-MXFP4 DSpark down projections.",
+      "source": "/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731",
+      "source_identity": {
+        "repo_id": "deepseek-ai/DeepSeek-V4-Flash-0731",
+        "revision": "9e165c30e2704aec5d9d593cce3eebd58bbef1cb",
+        "identity_cache": ".cache/huggingface/download"
+      },
+      "output": "DeepSeek-V4-Flash-0731-DS4-Quality128",
+      "ds4_revision": "d516d4eeb82c454aeb2831af1b1961801d6b571b",
+      "toolchain": {
+        "root": "/Users/andreas/code/3rd-party-repos/ds4-apetersson",
+        "upstream_base_revision": "4893e0c40fba03dbc85555faeb035799aa04e0b6",
+        "quantizer_sha256": "f0a381f4ada808ea2afa740d964354fa327fc1235ba7cebf50874eb89fb97ac5",
+        "runtime_sha256": "5382bee60cb1c2d1e3274511c56099f88279684fbf152fa409d551711dbb9543"
+      },
+      "main": {
+        "experts": "iq2_xxs",
+        "routed_w2": "q2_k",
+        "attention_proj": "q8_0",
+        "shared": "q8_0",
+        "preserve_indexer_f16": true,
+        "tensor_types": [
+          "output.weight=q8_0"
+        ],
+        "native_mxfp4_layers": [
+          10,
+          14,
+          30,
+          34,
+          37,
+          38,
+          39,
+          40,
+          41,
+          42
+        ],
+        "native_mxfp4_layer_selection": "Antirez 37-42 seed plus cross-recipe sensitivity controls 10,14,30,34; source packed I8 codes and F8_E8M0 scales are preserved exactly.",
+        "estimated_bytes": 102826238912
+      },
+      "dspark": {
+        "experts": "iq2_xxs",
+        "routed_w2": "mxfp4",
+        "block_size": 5,
+        "target_layers": [
+          40,
+          41,
+          42
+        ],
+        "markov_rank": 256,
+        "noise_token_id": 128799,
+        "calibration": "target-layer proxy aliases"
+      },
+      "verification": {
+        "require_imatrix_strict": true,
+        "require_native_byte_compare": true,
+        "main": {
+          "bytes": 102826238912,
+          "tensors": 1328,
+          "tensor_types": {
+            "i32": 3,
+            "iq2_xxs": 66,
+            "q2_k": 33,
+            "mxfp4": 30,
+            "f16": 359,
+            "f32": 492,
+            "q8_0": 345
+          }
+        },
+        "dspark": {
+          "bytes": 7297737120,
+          "tensors": 81,
+          "tensor_types": {
+            "f32": 34,
+            "f16": 7,
+            "q8_0": 31,
+            "iq2_xxs": 6,
+            "mxfp4": 3
+          }
+        }
+      },
+      "calibration": {
+        "repo_id": "ox-ox/DeepSeek-V4-Flash-0731-GGUF",
+        "revision": "6d58a3a36030c3ccb969bb5759fc6ae08cd299f8",
+        "filename": "imatrix/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat",
+        "bytes": 450892654,
+        "sha256": "6fce7674df701de544e5d3351aab04e67602eddeafeb48cf70e77ebe47239eb4",
+        "coverage_policy": "exact 129 routed names and vector dimensions; strict mode checks only imatrix-consuming output types"
+      },
+      "template": {
+        "role": "metadata-tokenizer-tensor-order-only",
+        "repo_id": "antirez/deepseek-v4-gguf",
+        "revision": "1cd7b564460821938add0475a60b942c409295e0",
+        "filename": "DeepSeek-V4-Flash-IQ2XXS-w2Q2K-AProjQ8-SExpQ8-OutQ8-chat-v2-imatrix-0731.gguf",
+        "remote_bytes": 86720111488,
+        "lfs_sha256": "ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0",
+        "xet_hash": "7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac",
+        "header_sha256": "f0e1d5e8f3b008402aa6eb32cada3873dd926c8bf5e7d00d7788eec65f09dd6d"
+      },
+      "recipe_control": {
+        "repo_id": "antirez/deepseek-v4-gguf",
+        "revision": "1cd7b564460821938add0475a60b942c409295e0",
+        "dspark_filename": "DeepSeek-V4-Flash-DSpark-support.gguf",
+        "weights_used": false
+      }
+    }
+  },
+  "calibration": {
+    "repo_id": "ox-ox/DeepSeek-V4-Flash-0731-GGUF",
+    "filename": "imatrix/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat",
+    "revision": "6d58a3a36030c3ccb969bb5759fc6ae08cd299f8",
+    "bytes": 450892654,
+    "sha256": "6fce7674df701de544e5d3351aab04e67602eddeafeb48cf70e77ebe47239eb4",
+    "coverage_validation": {
+      "entries": 129,
+      "scope": "all routed gate/up/down tensors",
+      "policy": "exact-name-and-vector-length",
+      "global_ds4_imatrix_strict": true
+    }
+  },
+  "template": {
+    "repo_id": "antirez/deepseek-v4-gguf",
+    "filename": "DeepSeek-V4-Flash-IQ2XXS-w2Q2K-AProjQ8-SExpQ8-OutQ8-chat-v2-imatrix-0731.gguf",
+    "revision": "1cd7b564460821938add0475a60b942c409295e0",
+    "role": "metadata-tokenizer-tensor-order-and-shapes-only",
+    "weights_used": false,
+    "remote_bytes": 86720111488,
+    "lfs_sha256": "ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0",
+    "xet_hash": "7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac",
+    "downloaded_header_bytes": 67108864,
+    "header_sha256": "f0e1d5e8f3b008402aa6eb32cada3873dd926c8bf5e7d00d7788eec65f09dd6d",
+    "resolved": {
+      "repo_id": "antirez/deepseek-v4-gguf",
+      "filename": "DeepSeek-V4-Flash-IQ2XXS-w2Q2K-AProjQ8-SExpQ8-OutQ8-chat-v2-imatrix-0731.gguf",
+      "role": "metadata-tokenizer-tensor-order-and-shapes-only",
+      "weights_used": false,
+      "remote_bytes": 86720111488,
+      "lfs_sha256": "ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0",
+      "xet_hash": "7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac",
+      "resolved_etag": "7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac",
+      "range": "bytes=0-67108863",
+      "header_bytes": 67108864,
+      "header_sha256": "f0e1d5e8f3b008402aa6eb32cada3873dd926c8bf5e7d00d7788eec65f09dd6d"
+    }
+  },
+  "dspark_calibration_alias": {
+    "schema": "deepseek-model-tools.dspark-imatrix-alias.v1",
+    "function": "extend_dspark_legacy_imatrix",
+    "source": {
+      "repo_id": "ox-ox/DeepSeek-V4-Flash-0731-GGUF",
+      "revision": "6d58a3a36030c3ccb969bb5759fc6ae08cd299f8",
+      "filename": "imatrix/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat",
+      "bytes": 450892654,
+      "sha256": "6fce7674df701de544e5d3351aab04e67602eddeafeb48cf70e77ebe47239eb4"
+    },
+    "output_sha256": "689b446ed2e2657ebcb69a6516781ee0444e07641aeb6125233fb4e6fe7cbce3",
+    "mapping": [
+      {
+        "stage": 0,
+        "target_layer": 40,
+        "alias": "mtp.0.ffn_gate_exps.weight",
+        "source": "blk.40.ffn_gate_exps.weight"
+      },
+      {
+        "stage": 0,
+        "target_layer": 40,
+        "alias": "mtp.0.ffn_up_exps.weight",
+        "source": "blk.40.ffn_up_exps.weight"
+      },
+      {
+        "stage": 0,
+        "target_layer": 40,
+        "alias": "mtp.0.ffn_down_exps.weight",
+        "source": "blk.40.ffn_down_exps.weight"
+      },
+      {
+        "stage": 1,
+        "target_layer": 41,
+        "alias": "mtp.1.ffn_gate_exps.weight",
+        "source": "blk.41.ffn_gate_exps.weight"
+      },
+      {
+        "stage": 1,
+        "target_layer": 41,
+        "alias": "mtp.1.ffn_up_exps.weight",
+        "source": "blk.41.ffn_up_exps.weight"
+      },
+      {
+        "stage": 1,
+        "target_layer": 41,
+        "alias": "mtp.1.ffn_down_exps.weight",
+        "source": "blk.41.ffn_down_exps.weight"
+      },
+      {
+        "stage": 2,
+        "target_layer": 42,
+        "alias": "mtp.2.ffn_gate_exps.weight",
+        "source": "blk.42.ffn_gate_exps.weight"
+      },
+      {
+        "stage": 2,
+        "target_layer": 42,
+        "alias": "mtp.2.ffn_up_exps.weight",
+        "source": "blk.42.ffn_up_exps.weight"
+      },
+      {
+        "stage": 2,
+        "target_layer": 42,
+        "alias": "mtp.2.ffn_down_exps.weight",
+        "source": "blk.42.ffn_down_exps.weight"
+      }
+    ],
+    "tool": {
+      "repository_revision": "e24746463a0a3e79036dd9c0472deac6ce704f08",
+      "importer_sha256": "b39987ce5ce35564bb4053ec11265c0ff41be233595ae17a6e039318a52dd4da"
+    },
+    "coverage_validation": {
+      "entries": 138,
+      "trunk_entries": 129,
+      "dspark_entries": 9,
+      "coverage": "exact-name-and-vector-length"
+    }
+  },
+  "tools": {
+    "ds4_revision": "d516d4eeb82c454aeb2831af1b1961801d6b571b",
+    "ds4_root": "/Users/andreas/code/3rd-party-repos/ds4-apetersson",
+    "ds4_quantizer_sha256": "f0a381f4ada808ea2afa740d964354fa327fc1235ba7cebf50874eb89fb97ac5",
+    "ds4_runtime_sha256": "5382bee60cb1c2d1e3274511c56099f88279684fbf152fa409d551711dbb9543",
+    "omlx_revision": "76352ed2363e42b2146243463875756e683a640d",
+    "deepseek_model_tools_revision": "e24746463a0a3e79036dd9c0472deac6ce704f08",
+    "deepseek_model_tools_worktree": {
+      "dirty": true,
+      "status_sha256": "92f5beb73ac6d01920de0cbfb8cd5cd4d019c05296a8bb3a183902b6eb83f1dc",
+      "tracked_patch_sha256": "64dfc4b91c70163b4848cb404cc85dd539967d22d789a3aa04567503895a6312"
+    },
+    "build_driver_sha256": "cb1569b909de010b4aaa2a1b1b7ed8e277bcdf41e32585f1651005b41e81567f"
+  },
+  "commands": [
+    [
+      "/Users/andreas/code/3rd-party-repos/ds4-apetersson/gguf-tools/deepseek4-quantize",
+      "--hf",
+      "/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731",
+      "--template",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/inputs/antirez/ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0.header-only.sparse.gguf",
+      "--experts",
+      "iq2_xxs",
+      "--routed-w2",
+      "q2_k",
+      "--attention-proj",
+      "q8_0",
+      "--shared",
+      "q8_0",
+      "--threads",
+      "8",
+      "--imatrix",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/inputs/ox-ox/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat",
+      "--tensor-type",
+      "output.weight=q8_0",
+      "--tensor-type",
+      "blk.2.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.4.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.6.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.8.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.10.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.12.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.14.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.16.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.18.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.20.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.22.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.24.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.26.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.28.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.30.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.32.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.34.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.36.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.38.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.40.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.42.indexer.attn_q_b.weight=f16",
+      "--tensor-type",
+      "blk.10.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.10.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.10.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.14.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.14.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.14.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.30.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.30.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.30.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.34.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.34.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.34.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.37.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.37.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.37.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.38.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.38.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.38.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.39.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.39.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.39.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.40.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.40.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.40.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.41.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.41.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.41.ffn_down_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.42.ffn_gate_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.42.ffn_up_exps.weight=mxfp4",
+      "--tensor-type",
+      "blk.42.ffn_down_exps.weight=mxfp4",
+      "--imatrix-strict",
+      "--out",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/.DeepSeek-V4-Flash-0731-DS4-Quality128.building-20260801-161944-54912/DeepSeek-V4-Flash-0731-DS4-Quality128.gguf"
+    ],
+    [
+      "/Users/andreas/code/3rd-party-repos/ds4-apetersson/gguf-tools/deepseek4-quantize",
+      "--hf",
+      "/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731",
+      "--template",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/inputs/antirez/ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0.header-only.sparse.gguf",
+      "--dspark-support",
+      "--experts",
+      "iq2_xxs",
+      "--routed-w2",
+      "mxfp4",
+      "--attention-proj",
+      "q8_0",
+      "--shared",
+      "q8_0",
+      "--dspark-block-size",
+      "5",
+      "--dspark-markov-rank",
+      "256",
+      "--dspark-noise-token-id",
+      "128799",
+      "--dspark-target-layers",
+      "40,41,42",
+      "--threads",
+      "8",
+      "--imatrix",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/inputs/ox-ox/DeepSeek-V4-Flash-0731-dspark-alias.dat",
+      "--out",
+      "/Users/andreas/Library/Caches/deepseek-model-tools/build128/.DeepSeek-V4-Flash-0731-DS4-Quality128.building-20260801-161944-54912/DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf",
+      "--imatrix-strict"
+    ]
+  ],
+  "appledouble_excluded": true,
+  "installation": "internal-stage -> hash-copy to hidden same-volume incoming -> verify -> atomic directory exposure via renameatx_np(RENAME_EXCL); when that kernel no-replace primitive is unsupported (notably exFAT), use a cooperative exclusive sibling lock + destination lstat + same-parent renameat fallback that preserves atomic visibility and refuses every destination observed before rename"
+}

DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:393b807aa0a409f1c006f0e700bbb91a80d3de28daa56c222edf944250075c66
+size 7297737120

DeepSeek-V4-Flash-0731-DS4-Quality128.gguf ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:efcbf786154c2aec61511785d04a4c8d98ee42500b60ff8a810717fc0a69d1d3
+size 102826238912

PROVENANCE.md ADDED Viewed

	@@ -0,0 +1,19 @@

+# Provenance
+This artifact was regenerated from the local official original FP8 checkpoint only.
+No tensors were copied from any public GGUF or another quantized model.
+- Weight source: `/Volumes/Samsung_4TB/models/DeepSeek-V4-Flash-0731`
+- Official source: `deepseek-ai/DeepSeek-V4-Flash-0731`
+- Official revision: `9e165c30e2704aec5d9d593cce3eebd58bbef1cb`
+- Full source-shard hashes verified: `true`
+- Profile: `/Users/andreas/code/deepseek-model-tools/configs/dsv4-ds4-quality128-original.json`
+- Calibration: `ox-ox/DeepSeek-V4-Flash-0731-GGUF/imatrix/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat` (`6fce7674df701de544e5d3351aab04e67602eddeafeb48cf70e77ebe47239eb4`)
+- DSpark calibration proxy: `689b446ed2e2657ebcb69a6516781ee0444e07641aeb6125233fb4e6fe7cbce3` (stages 0/1/2 from target layers 40/41/42)
+- DS4 template, metadata only: `antirez/deepseek-v4-gguf/DeepSeek-V4-Flash-IQ2XXS-w2Q2K-AProjQ8-SExpQ8-OutQ8-chat-v2-imatrix-0731.gguf`
+- Antirez full-object LFS SHA-256: `ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0`
+- Antirez Xet object: `7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac`
+- Antirez 64 MiB header SHA-256: `f0e1d5e8f3b008402aa6eb32cada3873dd926c8bf5e7d00d7788eec65f09dd6d`
+- Engine: `ds4`
+- DSpark: `true`
+- Profile description: Quality-first 128 GB DS4 profile for the official original DeepSeek V4 Flash 0731 checkpoint, preserving exact native MXFP4 routed experts on ten sensitivity-selected layers, with IQ2_XXS/Q2_K elsewhere and native-MXFP4 DSpark down projections.

README.md ADDED Viewed

	@@ -0,0 +1,379 @@

+---
+license: mit
+library_name: gguf
+pipeline_tag: text-generation
+base_model: deepseek-ai/DeepSeek-V4-Flash-0731
+base_model_relation: quantized
+quantized_by: apetersson
+tags:
+  - gguf
+  - quantized
+  - deepseek
+  - deepseek-v4
+  - deepseek-v4-flash
+  - moe
+  - mixture-of-experts
+  - mxfp4
+  - iq2_xxs
+  - q2_k
+  - ds4
+  - dspark
+  - apple-silicon
+  - metal
+---
+# DeepSeek V4 Flash 0731 — DS4 Quality128
+> **Clean official weights, exact MXFP4 experts, maximum resident quality.**
+> **Required DS4 branch:** this model is not compatible with an older or stock
+> DS4 build. Its native MXFP4 tensors require the official
+> [`antirez/ds4` `ds4f-mxfp4` branch](https://github.com/antirez/ds4/tree/ds4f-mxfp4).
+> That branch runs both the target model and the supplied DSpark support model.
+> **Validation status:** conversion and CPU structural validation are complete.
+> Metal generation, DSpark acceptance, throughput, peak-memory and actual
+> one-million-token-context tests for this exact artifact are still pending.
+This is a quality-first DS4 package of the official
+[`deepseek-ai/DeepSeek-V4-Flash-0731`](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731)
+checkpoint. It is designed to keep the target model—and optionally its real
+three-stage DSpark speculative drafter—resident on a 128 GB Apple Silicon host.
+This model was quantized directly from the original official checkpoint. No
+behavioral weight edit was applied before quantization. All model weights were
+regenerated from the official local FP8 checkpoint; no tensor values were
+copied from another GGUF or quantized model.
+## Artifacts
+| File | Purpose | Bytes | GiB | SHA-256 |
+| --- | --- | ---: | ---: | --- |
+| `DeepSeek-V4-Flash-0731-DS4-Quality128.gguf` | Authoritative 43-layer target model | `102,826,238,912` | `95.7644` | `efcbf786154c2aec61511785d04a4c8d98ee42500b60ff8a810717fc0a69d1d3` |
+| `DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf` | Three-stage speculative drafter; not standalone | `7,297,737,120` | `6.7965` | `393b807aa0a409f1c006f0e700bbb91a80d3de28daa56c222edf944250075c66` |
+| **Combined GGUFs** | — | **`110,123,976,032`** | **`102.5609`** | — |
+`BUILD_MANIFEST.json`, `PROVENANCE.md` and `SHA256SUMS` provide the full
+machine-readable build record and integrity inventory.
+## Quantization profile
+### Main target model
+| Tensor class | Quantization/storage | Rationale |
+| --- | --- | --- |
+| Routed gate/up/down on layers 10, 14, 30, 34, 37, 38, 39, 40, 41 and 42 | Exact native `MXFP4` | Preserves source packed I8 codes and F8_E8M0 scales byte-for-byte on sensitivity-selected layers. |
+| Routed gate/up on the other 33 MoE layers | `IQ2_XXS`, importance-matrix calibrated | Applies the most aggressive compression to the largest tensor bank. |
+| Routed down on the other 33 MoE layers | `Q2_K`, importance-matrix calibrated | More conservative two-bit storage on the projection that writes expert output to the residual stream. |
+| Attention projections | `Q8_0` | Protects a dense path used for every token. |
+| Shared experts | `Q8_0` | Protects the expert path active for every token. |
+| Vocabulary/output head | `Q8_0` | Protects final-logit fidelity. |
+| Indexer `attn_q_b` tensors on layers 2, 4, …, 42 | forced `F16` | Preserves a small, sensitive compressed-attention component. |
+| Remaining control, normalization, routing and auxiliary tensors | template-declared `F16`/`F32`/`I32` | Avoids forcing small or numerically sensitive tensors into the low-bit expert rules. |
+Observed main-GGUF inventory from strict DS4 inspection:
+| GGUF type | Tensors |
+| --- | ---: |
+| `F32` | 492 |
+| `F16` | 359 |
+| `I32` | 3 |
+| `Q8_0` | 345 |
+| `IQ2_XXS` | 66 |
+| `Q2_K` | 33 |
+| `MXFP4` | 30 |
+| **Total** | **1,328** |
+The target GGUF is version 3, describes approximately 284.33 billion logical
+parameters and retains the checkpoint's declared 1,048,576-token training
+context. That declaration is not evidence that a one-million-token inference
+run fits or remains robust on a particular machine.
+### DSpark support model
+This is the actual 0731 three-stage DSpark module (`mtp.0`–`mtp.2`), not the
+legacy single-stage MTP attachment. The target model remains authoritative and
+verifies speculative proposals.
+| Parameter | Value |
+| --- | --- |
+| Stages | 3 |
+| Proposal block size | 5 |
+| Target layers | 40, 41, 42 |
+| Markov rank | 256 |
+| Noise token ID | 128799 |
+| Routed gate/up | `IQ2_XXS` (6 tensors) |
+| Routed down | exact native `MXFP4` (3 tensors) |
+| Dense projections | `Q8_0` (31 tensors) |
+| Control/auxiliary tensors | 7 `F16` + 34 `F32` tensors |
+| Total | 81 tensors |
+The support GGUF describes approximately 19.85 billion logical parameters and
+must be loaded alongside its matching target GGUF.
+## Importance calibration
+The routed-expert calibration source is the 0731-native matrix from
+[`ox-ox/DeepSeek-V4-Flash-0731-GGUF`](https://huggingface.co/ox-ox/DeepSeek-V4-Flash-0731-GGUF):
+- Repository revision: `6d58a3a36030c3ccb969bb5759fc6ae08cd299f8`
+- File: `imatrix/DeepSeek-V4-Flash-0731-chat-v2-routed-moe-ds4-1p5m.dat`
+- Size: `450,892,654` bytes
+- SHA-256: `6fce7674df701de544e5d3351aab04e67602eddeafeb48cf70e77ebe47239eb4`
+- Coverage: exactly 129 routed target tensors (43 layers × gate/up/down), with exact name and vector-dimension validation
+The build ran strict imatrix validation. Preserved native MXFP4 tensors do not
+consume imatrix data because they are copied from the official FP8 checkpoint's
+packed source representation rather than requantized.
+The published matrix has no native DSpark entries. The support-model importer
+therefore made deterministic target-layer proxy aliases:
+- `mtp.0` ← target layer 40
+- `mtp.1` ← target layer 41
+- `mtp.2` ← target layer 42
+The extended matrix contains 138 entries and has SHA-256
+`689b446ed2e2657ebcb69a6516781ee0444e07641aeb6125233fb4e6fe7cbce3`.
+This is an explicitly recorded proxy, not a fresh activation capture from the
+DSpark drafter.
+## Weight provenance and metadata template
+The sole source of weight values was the local copy of
+`deepseek-ai/DeepSeek-V4-Flash-0731` at official revision
+`9e165c30e2704aec5d9d593cce3eebd58bbef1cb`. All 48 FP8 weight shards—
+`166,886,535,336` bytes in total—were fully SHA-256 verified and guarded
+against mutation throughout conversion.
+The exact-0731 GGUF from
+[`antirez/deepseek-v4-gguf`](https://huggingface.co/antirez/deepseek-v4-gguf)
+was used only as a bounded metadata, tokenizer, tensor-order and shape template:
+- Repository revision: `1cd7b564460821938add0475a60b942c409295e0`
+- Template LFS SHA-256: `ca22ae2f838e14077c22bc1c1417b71b45b5e5a3687bd96c2ac6e17fdb6261c0`
+- Template Xet object: `7da16e1025c1b856490c29c341f4e467d15cb195389c70383dade5e6108799ac`
+- Verified 64 MiB header SHA-256: `f0e1d5e8f3b008402aa6eb32cada3873dd926c8bf5e7d00d7788eec65f09dd6d`
+- Template weight values used: **no**
+## Runtime requirement
+The required upstream runtime line is the official
+[`antirez/ds4` `ds4f-mxfp4` branch](https://github.com/antirez/ds4/tree/ds4f-mxfp4),
+not Antirez's default branch and not an older DS4 release. This is required
+because the main GGUF contains 30 native MXFP4 routed-expert tensors. A runtime
+without that branch's MXFP4 loader and Metal kernels is not compatible, even if
+it can parse the GGUF header.
+The upstream branch revision used as the base for this build was
+`4893e0c40fba03dbc85555faeb035799aa04e0b6`. The required runtime checkout is:
+```bash
+git clone --branch ds4f-mxfp4 https://github.com/antirez/ds4.git
+cd ds4
+make
+```
+Build note: the [`d516d4e` quantizer PR](https://github.com/apetersson/ds4-omlx/tree/d516d4eeb82c454aeb2831af1b1961801d6b571b),
+tracked as [`antirez/ds4#642`](https://github.com/antirez/ds4/issues/642),
+was needed to create the preserved-MXFP4 DSpark GGUF. Users do **not** need that
+PR to run either supplied file.
+Do not substitute `antirez/ds4` `main`, an older DS4 binary or a generic GGUF
+runtime. Container parsing alone does not demonstrate correct native MXFP4 or
+mixed DSpark execution.
+## Running with DS4
+Target-only Metal inference:
+```bash
+./ds4 --metal \
+  -m DeepSeek-V4-Flash-0731-DS4-Quality128.gguf
+```
+Greedy DSpark inference:
+```bash
+./ds4 --metal \
+  -m DeepSeek-V4-Flash-0731-DS4-Quality128.gguf \
+  --mtp DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf \
+  --dspark --temp 0
+```
+DSpark is opt-in, accelerates generation rather than prefill, and can be neutral
+or slower when proposal acceptance is low. Sampled decoding does not use DSpark
+proposals in the pinned runtime.
+Structural inspection:
+```bash
+./ds4 --cpu --inspect \
+  -m DeepSeek-V4-Flash-0731-DS4-Quality128.gguf \
+  --mtp DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf \
+  --dspark-strict
+```
+## Memory planning on a 128 GB Mac
+The target machine has 128 GiB (`137,438,953,472` bytes) of unified physical
+memory, but the more relevant ceiling for a large Metal allocation is its
+approximately `121.60 GiB` `recommendedMaxWorkingSetSize`. The remaining
+roughly `6.40 GiB` is not extra model capacity; macOS, applications, drivers and
+untracked transient allocations still need memory. A configuration at or below
+128 GiB can therefore be unsafe when it is above—or too close to—the Metal
+working-set recommendation.
+### Fixed weight cost
+| Loaded weights | GiB | Share of 128 GiB | Share of 121.60 GiB Metal recommendation |
+| --- | ---: | ---: | ---: |
+| Target model only | `95.7644` | `74.8%` | `78.8%` |
+| Target + DSpark support | `102.5609` | `80.1%` | `84.3%` |
+These are GGUF payload sizes, before KV cache, indexed-attention scratch,
+prefill workspace, verifier state and other runtime allocations. DS4's planner
+uses an approximately `97.63 GiB` resident span for the main model after its
+mapping/alignment accounting, rather than treating the main file size as the
+entire live allocation.
+### Context-length scaling at prefill chunk 1,024
+`--ctx` is the total prompt-plus-completion capacity. `--prefill-chunk` is the
+maximum prompt microbatch DS4 processes at once; it is the relevant “batch
+size” for this single-session memory calculation. A larger chunk can improve
+prefill throughput, but indexed-attention scratch grows with both context length
+and chunk size.
+The following are conservative planning estimates. “Margin” is remaining space
+under the `121.60 GiB` Metal recommendation, not free system RAM.
+| Context | DSpark off: total | Off: margin | DSpark on: total | On: margin |
+| ---: | ---: | ---: | ---: | ---: |
+| 4,096 | `97.77 GiB` | `23.83 GiB` | `104.57 GiB` | `17.03 GiB` |
+| 32,768 | `98.05 GiB` | `23.55 GiB` | `104.85 GiB` | `16.75 GiB` |
+| 131,072 | `98.99 GiB` | `22.61 GiB` | `105.79 GiB` | `15.81 GiB` |
+| 262,144 | `100.24 GiB` | `21.36 GiB` | `107.04 GiB` | `14.56 GiB` |
+| 524,288 | `102.75 GiB` | `18.85 GiB` | `109.55 GiB` | `12.05 GiB` |
+| 1,048,576 | `107.77 GiB` | `13.83 GiB` | `114.56 GiB` | `7.04 GiB` |
+At ordinary 4K–128K contexts, the model weights dominate and chunk 1,024 leaves
+substantial planned margin. At 512K and especially 1M, the compressed KV and
+context-by-chunk attention workspace become material. DSpark adds approximately
+`6.80 GiB` at every context length because its support model remains resident
+during prefill even though speculative decoding only accelerates generation.
+### Prefill batch-size effect at one-million-token context
+This table holds `--ctx 1048576` constant and varies `--prefill-chunk`. Totals
+are the conservative envelope of DS4's resident-span/context estimator and the
+completed release planning model. Percentages use all 128 GiB of physical RAM;
+the Metal margin remains the safer operational measure.
+| Prefill chunk | DSpark off total | Off: 128 GiB used | Off: Metal margin | DSpark on total | On: 128 GiB used | On: Metal margin |
+| ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| 256 | `106.20 GiB` | `83.0%` | `15.40 GiB` | `113.00 GiB` | `88.3%` | `8.60 GiB` |
+| 512 | `106.72 GiB` | `83.4%` | `14.88 GiB` | `113.52 GiB` | `88.7%` | `8.08 GiB` |
+| 1,024 | `107.77 GiB` | `84.2%` | `13.83 GiB` | `114.56 GiB` | `89.5%` | `7.04 GiB` |
+| 2,048 | `110.30 GiB` | `86.2%` | `11.30 GiB` | `117.09 GiB` | `91.5%` | `4.51 GiB` |
+| 4,096 | `116.44 GiB` | `91.0%` | `5.16 GiB` | `123.24 GiB` | `96.3%` | **`−1.64 GiB`** |
+The 4,096/DSpark combination exceeds Metal's recommendation despite fitting
+numerically inside 128 GiB and should not be treated as resident-safe. The
+2,048/DSpark combination is also tight: its `4.51 GiB` planned Metal margin can
+be consumed by omitted driver and verifier peaks. Chunk 1,024 is the sensible
+first DSpark experiment at 1M; target-only chunk 2,048 is the more conservative
+maximum-quality starting point. Chunks 256–512 provide more margin at the cost
+of more prefill iterations and likely lower prompt-processing throughput.
+Target-only 1M starting point:
+```bash
+./ds4 --metal \
+  -m DeepSeek-V4-Flash-0731-DS4-Quality128.gguf \
+  --ctx 1048576 --prefill-chunk 2048
+```
+DSpark 1M starting point, only after unloading other large applications and
+measuring the target-only peak:
+```bash
+./ds4 --metal \
+  -m DeepSeek-V4-Flash-0731-DS4-Quality128.gguf \
+  --ctx 1048576 --prefill-chunk 1024 \
+  --mtp DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf \
+  --dspark --temp 0
+```
+### Multiple sessions and server batching
+The tables describe one resident session. They do not mean that `N` concurrent
+requests cost exactly `N` times the displayed total: weights and some graph
+workspace are shared, while each resident session needs its own KV/context
+buffers and live request state. DS4 reports the aggregate minimum context-buffer
+request at server startup. For multi-user serving, start with one session,
+measure the process and system peak, then raise concurrency one session at a
+time; do not infer a safe concurrency count from the GGUF sizes alone.
+All figures above are planning estimates, not measurements of this exact
+artifact. Metal load, peak-memory capture and an actual 1,048,576-token run
+remain required. Keep several GiB of additional operational margin, watch
+memory pressure rather than only Activity Monitor's process RSS, and expect
+other active models or large applications to invalidate the table.
+## Reproducibility
+- Build tooling revision: `e24746463a0a3e79036dd9c0472deac6ce704f08`
+- Build driver SHA-256: `cb1569b909de010b4aaa2a1b1b7ed8e277bcdf41e32585f1651005b41e81567f`
+- Build-time DS4/quantizer revision: `d516d4eeb82c454aeb2831af1b1961801d6b571b`
+- oMLX revision: `76352ed2363e42b2146243463875756e683a640d`
+- Profile SHA-256: `bc7f6470a0cb6208addf2ac52806d976108517119b411023da3fd91b492b218c`
+- Build run ID: `20260801-161944-54912`
+The tooling worktree was intentionally dirty and is cryptographically described
+in `BUILD_MANIFEST.json`, alongside the complete profile, source-shard hashes,
+calibration/template identities and exact converter commands.
+## Validation status
+Completed on 2026-08-01:
+- focused build-driver and finalization test suites: 40/40 passed;
+- every one of the 48 official source weight shards fully SHA-256 verified;
+- strict routed-imatrix name and vector-dimension coverage passed;
+- main GGUF: exact size, 1,328 tensors, exact name set and exact type histogram;
+- DSpark GGUF: exact size, 81 tensors, three stages and target layers 40/41/42;
+- strict DSpark binding: 81 tensors, 0 missing, 0 invalid and 0 metadata errors;
+- `ds4 --cpu --inspect --dspark-strict` passed;
+- byte-for-byte regeneration passed for all 30 main and all three DSpark native
+  MXFP4 tensors; and
+- the original build payloads passed their recorded SHA-256 checks before
+  atomic publication; this README was added afterward and independently added
+  to `SHA256SUMS`.
+Still required before making runtime, quality or performance claims:
+- successful Metal load and deterministic generation with DSpark disabled;
+- successful Metal generation with DSpark enabled and lossless target agreement;
+- proposal acceptance rate and accepted tokens per target step;
+- prompt-processing and generation tokens/second;
+- measured peak unified memory and practical context limits on the target host;
+- an actual 1,048,576-token context run; and
+- capability/perplexity comparisons against the official FP8 source.
+## Limitations and responsible use
+Ultra-low-bit expert quantization can reduce reasoning, factuality, style
+fidelity and long-context robustness even when dense paths and selected experts
+are protected. Native MXFP4 Metal kernels and the mixed DSpark path are newer
+than DS4's mature Q4_K path. This package does not guarantee correctness,
+neutrality, safety, regulatory compliance or a particular response style.
+Evaluate it for the intended workload and apply appropriate access controls.
+## License and attribution
+The upstream repository and weights are MIT licensed. This quantized derivative
+retains that license. Credit DeepSeek-AI for the original model, ox-ox for the
+0731 routed-expert importance matrix, antirez for DS4 and the exact-0731
+metadata recipe, and apetersson for the DS4 fork, conversion profile and release
+tooling.

SHA256SUMS ADDED Viewed

	@@ -0,0 +1,5 @@

+06c21a905c65f01394583d46e1f120fffabdae7dbaae610903e1ffb441ac720d  BUILD_MANIFEST.json
+393b807aa0a409f1c006f0e700bbb91a80d3de28daa56c222edf944250075c66  DeepSeek-V4-Flash-0731-DS4-Quality128-DSpark-support.gguf
+efcbf786154c2aec61511785d04a4c8d98ee42500b60ff8a810717fc0a69d1d3  DeepSeek-V4-Flash-0731-DS4-Quality128.gguf
+05546a9a7511cae87b0eb34191654bba23c6746e134c3b0c6e7d61a03909f1e0  PROVENANCE.md
+93b97d6c9bea07f26936b74ba5c0f09ef4d4938789c259844f201410cbf3038e  README.md