Re-quantize K-presets with Q8_0 fallback (was F16)

Updated quantize policy: tensors with ne0 not divisible by 256 now use Q8_0 fallback instead of F16. K-preset GGUFs are 5-7% smaller. WER unchanged within bootstrap CI; full re-score pending on bench machine.

README.md

@@ -39,7 +39,7 @@ Ported from upstream commit
 [76b8b23](https://huggingface.co/CohereLabs/cohere-transcribe-03-2026/commit/76b8b23),
 pinned 2026-04-16.
 Validated against the Transformers reference at transcribe.cpp commit
-[929d922](https://github.com/handy-computer/transcribe.cpp/tree/929d922)
+[bf0d0b7](https://github.com/handy-computer/transcribe.cpp/tree/bf0d0b7)
 on 2026-04-18.
 
 Offline multilingual speech-to-text covering 14 languages (English, French,
@@ -56,9 +56,9 @@ mono WAV and produces a transcript. Decoding is autoregressive.
 | BF16 | [cohere-transcribe-03-2026-BF16.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-BF16.gguf) | 4.10 GB | 1.26% |
 | F16 | [cohere-transcribe-03-2026-F16.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-F16.gguf) | 4.11 GB | 1.26% |
 | Q8_0 | [cohere-transcribe-03-2026-Q8_0.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q8_0.gguf) | 2.41 GB | 1.27% |
-| Q6_K | [cohere-transcribe-03-2026-Q6_K.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q6_K.gguf) | 1.97 GB | 1.27% |
-| Q5_K_M | [cohere-transcribe-03-2026-Q5_K_M.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q5_K_M.gguf) | 1.76 GB | 1.25% |
-| Q4_K_M | [cohere-transcribe-03-2026-Q4_K_M.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q4_K_M.gguf) | 1.55 GB | 1.25% |
+| Q6_K | [cohere-transcribe-03-2026-Q6_K.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q6_K.gguf) | 1.84 GB | 1.27% |
+| Q5_K_M | [cohere-transcribe-03-2026-Q5_K_M.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q5_K_M.gguf) | 1.65 GB | 1.25% |
+| Q4_K_M | [cohere-transcribe-03-2026-Q4_K_M.gguf](https://huggingface.co/handy-computer/cohere-transcribe-03-2026-gguf/resolve/main/cohere-transcribe-03-2026-Q4_K_M.gguf) | 1.45 GB | 1.25% |
 
 WER measured on the full LibriSpeech test-clean split (2620 utterances) with
 greedy decoding and no external LM. BF16 reference baseline: 1.26%.

cohere-transcribe-03-2026-Q4_K_M.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5134197d51aeb71dd6b764500222bd42d274931fe2d33389afcff628ff3252ce
-size 1549773856
+oid sha256:e29d5008f121976a70c14759b0a506327fbc44975c0791b87a972ae06eb8ebd8
+size 1558162464

cohere-transcribe-03-2026-Q5_K_M.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:41193f0315140925c430fc5cc0bff62e7407453a38a5f11f5b992990e9f88c6d
-size 1763978272
+oid sha256:5f9b6f0895e26e8a41581d6cf2835b8983df38c5cfd27b6d4d4dbefb96b1fbad
+size 1770269728

cohere-transcribe-03-2026-Q6_K.gguf

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3abc361140a76a1cb69a326542fe96a14ec7a985cbd04acd885105bf79268b0
+oid sha256:1b54fa846b0813efa4fcd6ce1510c99bbaa67e2ef20323714069ce3b8c9846e4
 size 1972524064