Instructions to use Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16")
model = AutoModelForCausalLM.from_pretrained("Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16

SGLang

How to use Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 with Docker Model Runner:
```
docker model run hf.co/Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16
```

Ornith-1.5-35B-A3B REAP-50 NVFP4A16 (16 GB)

REAP expert-pruned (50%) + GPTQ-NVFP4A16 quantized build of ornith-ai/Ornith-1.5-35B-A3B, sized and served to run as a local coding model inside 16 GB of consumer VRAM — 12.47 GiB, RTX 5070 Ti (SM120), vLLM. Text-only: the base model's genuinely-trained vision tower and its MTP draft head are deliberately removed (documented trade-offs below), which is what makes the text-only architecture (Qwen3_5MoeForCausalLM) servable by stock-class vLLM after two confirmed upstream gaps were patched locally (both independently fixed upstream since, in vLLM #50210).

First quality signal — HumanEval+ 84.15% [77.8, 88.9], HumanEval 90.24% [84.7, 93.9], MBPP+ 89.15% [85.6, 91.9], greedy decoding, instruct framing, measured on exactly the released checkpoint. SWE-bench Verified has not been measured yet for this build; see the evaluation section for how it will be — this card will be updated when those numbers exist, and no agentic claim should be inferred from the code benchmarks above.

Highlights

Result	Detail
12.47 GiB	REAP 50% expert pruning (256→128 experts) + GPTQ-NVFP4A16 (weight-only), vision tower and MTP head stripped
84.15% [77.8, 88.9]	HumanEval+, greedy, instruct framing, n=164 — first genuine accuracy measurement of this pipeline, on the released artifact
89.15% [85.6, 91.9]	MBPP+, same protocol, n=378
~19 B params (from 35.95 B)	Expert halving is what pays for the 16 GB budget; 8-of-128 expert routing preserved
MIT	Inherited from the base model

Why 50 percent

Forced by arithmetic on a 16 GB card, not a tuning choice:

variant	size	fits 16 GB
bf16 base (measured footprint)	~65.4 GiB	no
NVFP4, unpruned (`ornith-ai/Ornith-1.5-35B-A3B-NVFP4`)	~21.8 GiB	no — before any KV cache
REAP 25% + NVFP4	not built	KV-cache headroom at 25% was insufficient on the prior project's identical-budget build; not re-derived here
REAP 50% + NVFP4A16	12.47 GiB	yes, with KV cache room

Supporting evidence for the ratio: Half the Experts, All the Code pruned Qwen3.6-35B-A3B — this base model's size-class cousin — at 50% with no statistically detectable loss on its primary code benchmark.

Evaluation

HumanEval+ and MBPP+ via lm-eval-harness, greedy decoding, instruct framing, Wilson 95% confidence intervals, served through vLLM on the released checkpoint itself:

benchmark	score	95% CI	n
HumanEval+	84.15%	[77.8, 88.9]	164
HumanEval	90.24%	[84.7, 93.9]	164
MBPP+	89.15%	[85.6, 91.9]	378

The full 706-problem suite ran in ~12 minutes through vLLM's backend. For tier context, our previous release (KAT-Coder-V2.5-Dev REAP-50 NVFP4A16, a different and code-specialized base) scored 96.34 / 89.63 / 89.42 on the same three tasks: this build is lower across the board but in the same tier, consistent with more aggressive pruning (same 50%) applied to a general-purpose MoE base rather than a code-specialized one. No upstream HumanEval/MBPP numbers are published for Ornith-1.5-35B-A3B itself, so there is no published figure to compare against directly.

SWE-bench Verified — not yet measured

This project publishes SWE-bench Verified only after a fixed validation ladder: single-instance smoke → small bounded sample → full 50-instance pilot, with the same promotion discipline as the prior release. That ladder has not started for this checkpoint. When it completes, this section gets the score, the full failure-mode breakdown, and the context-ceiling caveats the prior release's card documents. Until then: no agentic performance claim is made or implied.

Quantization and pruning details

Field	Value
Base model	`ornith-ai/Ornith-1.5-35B-A3B` (35.95 B params, MIT)
Pruning	REAP, expert-level, 50% compression ratio (256→128 experts), seed 42, single seed
Pruning residency	`--residency cpu_full` — validated deterministic path; a `layerwise` detour surfaced three real `reap-cuda` bugs and was reverted after the RAM math that motivated it was re-measured and found overly conservative (~65.4 GiB actual vs ~71.9 GiB estimated)
Pruning calibration	`theblackcat102/evol-codealpaca-v1`
Router renormalization	Fixed (upstream REAP adapter silently disables it for this architecture; fix carried from the prior release)
MTP head	Removed (`mtp_num_hidden_layers: 0`). Confirmed expert-routed — a full extra decoder layer with its own 256-expert MoE — meaning joint pruning was unvalidated; disabling it forfeits speculative-decoding speedups, documented as accepted trade-off
Vision tower	Removed. Verified genuinely trained (not phantom weights) by statistical comparison against known-init fingerprints; removal is a deliberate capability trade-off, not dead-weight stripping. Checkpoint switched to transformers' real text-only class (`Qwen3_5MoeForCausalLM`)
Quantization method	compressed-tensors / llm-compressor 0.13.0, GPTQ rounding (chosen over RTN from the start — the prior project's "GPTQ≈RTN" finding doesn't transfer to this expert topology)
Quantization scheme	NVFP4A16 — weight-only 4-bit float, group size 16, fp8_e4m3 scales, static actorder, tensor_group strategy; activations bf16
Quantization calibration	evol-codealpaca, 256 samples, 2048 sequence length, all experts calibrated
Kept unquantized	291 tensors: routers, shared-expert gates, embeddings, linear-attention projections, lm_head
Files	Single `model.safetensors` (12.47 GiB) plus tokenizer/config
Serving validation	Two confirmed upstream gaps were required to serve this checkpoint and were patched locally; both have since been fixed upstream independently (vLLM #50210): a missing model registry entry for `Qwen3_5MoeForCausalLM`, and three `IsHybrid`-required `get_mamba_state_*` classmethods missing from `Qwen3_5ForCausalLMBase`. The pinned, locally-patched vLLM tree that produced the numbers below is referenced in the eval scripts. Separately, loading this checkpoint family through stock llm-compressor ≤0.13.0 silently drops every quantized expert weight (no `qwen3_5_moe` entry in `ARCH_TO_2D_MAPPINGS`) — if you load these weights outside vLLM, verify your loader keeps them; upstream fix tracked in llm-compressor PR #3080 (approved, unmerged as of 2026-08-25)
Built on	RTX 5070 Ti, 16 GB VRAM, SM120 (compute capability 12.0)

Usage

Requires vLLM with SM120 support. The configuration below is the one actually exercised end-to-end during evaluation (via vLLM's engine, greedy):

vllm serve Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16 \
  --dtype bfloat16 --trust-remote-code \
  --gpu-memory-utilization 0.90 \
  --max-num-seqs 8 \
  --max-cudagraph-capture-size 8 \
  --mamba-cache-mode align \
  --enable-prefix-caching \
  --mamba-block-size 528 --block-size 528

Notes, each earned the hard way:

--max-cudagraph-capture-size below vLLM's default of 512 avoids a documented mamba-cache assertion for this architecture family; 8 is what the evaluated configuration used.
--mamba-block-size 528 --block-size 528 alignment is required for correct hybrid-cache sizing on this checkpoint.
Prefix caching works, but is ineffective for prompts shorter than the 528-token mamba block boundary (open upstream issue vllm-project/vllm #40696) — expect no prefix-cache benefit on short early turns in agentic workloads.
max_model_len was capped at 2048 during benchmarking; larger contexts are expected to work but were not part of the measured configuration, and the safe ceiling on a 16 GB card will be published with the SWE-bench results.
Greedy (temperature=0) is what every number on this card was measured with. No sampling recommendation is made yet; sampling sweeps belong to the SWE-bench phase.

Known limitations

No pruning-ablation baseline measured. The unpruned model does not fit this hardware; the accuracy cost of REAP itself (independent of quantization) is not isolated here. The comparison against the prior release above is tier context, not an ablation.
Text-only by construction. The vision tower was trained and worked; this build cannot see images. Documented trade-off, reversible in principle by rebuilding from the unstripped variant.
No MTP / speculative decoding. Forfeited by the MTP-strip decision above.
Context ceiling unpublished. The measured evaluation ran at 2,048 tokens; the safe serving ceiling for 16 GB will be stated with the SWE-bench results rather than guessed here.
These are self-reported numbers with published reproduction scripts, independently re-runnable from the eval suite in the companion repository; they are not leaderboard submissions.

Prior art and scope of claims

Verified against the Hugging Face Hub on 2026-08-25:

Unpruned NVFP4 of this base exists officially: ornith-ai/Ornith-1.5-35B-A3B-NVFP4 (~21.8 GiB).
REAP-pruned Ornith builds did not exist on the Hub as of that date.

What is distinct, and all that is claimed: a vLLM-servable Ornith that fits 16 GB with KV-cache room, with published HumanEval+/MBPP+ numbers, Wilson intervals, and the exact serving configuration — none of which the unpruned quant above publishes. The upstream-gap fixes required to serve text-only Qwen3.5-MoE checkpoints have landed upstream (vLLM #50210), so current vLLM builds serve this architecture without local patching.

License

MIT, inherited from the base model ornith-ai/Ornith-1.5-35B-A3B.

Citation

This checkpoint is derived from ornith-ai/Ornith-1.5-35B-A3B. If you use it, please cite the base model:

@misc{ornith15_2026,
  title={Ornith-1.5-35B-A3B},
  author={{Ornith AI}},
  year={2026},
  url={https://huggingface.co/ornith-ai/Ornith-1.5-35B-A3B}
}

And the pruning method:

@misc{reap2025,
  title={{REAP} the Experts: Why Pruning Prevails for One-Shot {MoE} compression},
  author={Lasby, Mike and Lazarevich, Ivan and Sinnadurai, Nish and Lie, Sean and Ioannou, Yani and Thangarasa, Vithursan},
  year={2025},
  eprint={2510.13999},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2510.13999}
}

Downloads last month: -

Safetensors

Model size

19B params

Tensor type

BF16

Model tree for Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16

Base model

ornith-ai/Ornith-1.5-35B-A3B

Quantized

(91)

this model

Paper for Ttimms/Ornith-1.5-35B-A3B-REAP-50-NVFP4A16

REAP the Experts: Why Pruning Prevails for One-Shot MoE compression

Paper • 2510.13999 • Published Oct 15, 2025 • 21