mirror of
https://github.com/ollama/ollama.git
synced 2026-09-21 05:28:00 -05:00
Includes quantized matmul corruption fix, which impacts nvfp4 multimodal models (gemma4 vision towers). Deferring wiring up the new MLX-C thread-local stream/sync APIs for now.
1 line
40 B
Plaintext
1 line
40 B
Plaintext
ebc88f10caa1b625e6b581437a8dea6df8a70085 |