The source configuration's 1M-token YaRN settings are preserved, but this release was locally exercised only through 19,069 input tokens—not at 1M.
This variant packages the complete MXFP4 body and vision tower with the source model's native 15-tensor BF16 MTP head for MTPLX 2.0.1. Depth 2 is the recommended setting.
1mtplx inspect --require-mtp <local-path-or-repo-id>
2mtplx serve --model <local-path-or-repo-id> --generation-mode mtp --mtp-depth 2 --reasoning-mode off
Local 192-token tuning measured 44.96 decode tok/s at depth 2 versus 25.39 autoregressive (1.77x). In a deterministic 10-prompt API comparison, depth 2 and autoregressive outputs were byte-identical on all 10 prompts; both passed the same 6/10 validators. Mean decode throughput was 49.45 versus 27.87 tok/s. A separate code smoke passed 3/3. Thermal fan control remained on auto.
1python -m mlx_vlm.generate \
2 --model <local-path-or-repo-id> \
3 --image image.jpg \
4 --prompt "Describe this image." \
5 --max-tokens 256
Review the
source model card for intended use, training, license, and deployment considerations. This conversion does not change the source model's safety characteristics.
