divinetribe/MiMo-V2.6-9B-MLX-bf16

MiMo-V2.6-Distill-Qwen-9B — Vision + Code (bf16, MLX)

Full-precision bf16 MLX conversion of Xiaomi's MiMo-V2.6-Distill-Qwen-9B, with the vision tower kept intact so it both reads images and writes code on Apple Silicon. No quantization: every weight is the original bf16. 760 tensors, 18.8 GB, four shards.

This is the full-quality build. Smaller sizes that passed the same vision check: 8-bit (about half the size, for 16 GB machines) · 4-bit (smallest, for 8 GB machines).

MiMo-V2.6-Distill-Qwen-9B is a 9B agentic model from the Xiaomi MiMo team, built by supervised fine-tuning of Qwen3.5-9B on MiMo-generated data covering coding, general agent tasks, visual coding, and cybersecurity.

What it is

• Base: Qwen/Qwen3.5-9B (Qwen team).
• Source checkpoint: XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B (MIT).
• Architecture: Qwen3_5ForConditionalGeneration — a qwen3_5_vision tower feeding the qwen3_5_text decoder. 9.4B parameters.
• This repo: bf16 MLX weights, vision kept, for mlx-vlm.
• Other sizes: 8-bit · 4-bit.

Running it (MLX)

from mlx_vlm import load, generate
from mlx_vlm.prompt_utils import apply_chat_template

model, processor = load("divinetribe/MiMo-V2.6-9B-MLX-bf16")
prompt = apply_chat_template(processor, model.config,
                             "What is in this image?", num_images=1)
print(generate(model, processor, prompt, image=["photo.jpg"], max_tokens=256))

Text-only works with num_images=0 and no image= argument.

Worked vision example — pose-match QC

A real frame from a motion game (a translucent target avatar drawn over a live camera scene) was fed to this converted bf16 model, asking whether the drawn figure is a correct human pose and whether any limbs are extra or wrong. The model's unedited on-device answer:

The figure is a single human silhouette with two legs and two arms, so there are no extra limbs or crossed knees. The legs stand apart in a wide stance with knees slightly bent and feet planted on the floor. The arms hang down at the sides with elbows slightly bent, hands near the hips. The pose looks anatomically correct and matches a neutral boxing stance.

This is grounded in the actual pixels: it reads the single figure, counts the limbs correctly (two arms, two legs, no 4-leg or crossed-knee artifact — the exact QC signal we want), gets the wide stance and bent knees right, and names the boxing context. On this frame it placed the arms low rather than up in a guard, so it is a genuine, imperfect QC read, shown unedited. The same vision path writes alt-text for product and UI images and reads screenshots back into code.

Benchmarks

The single most reputable public benchmark that genuinely applies here is SWE-bench Verified — a human-validated set of real GitHub issues with hidden tests, the standard yardstick for coding agents. In the MiMo-V2.6 technical report, this SFT checkpoint scores 61.1 (avg@3) on SWE-bench Verified, ahead of its Qwen3.5-9B base at 60.0. Other numbers in that report (MiMo Code, MiMo Visual Coding, etc.) are the authors' internal evaluation sets, so treat those as the authors' own rather than an external leaderboard. No scores here are invented — they are quoted from the source model card and the technical report.

Credits

Enormous thanks to the people whose work this conversion simply repackages:

• The Xiaomi MiMo team — they built and released MiMo-V2.6-Distill-Qwen-9B under MIT: model · org. Citation: MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement, Xiaomi MiMo Team, 2026.
• The Qwen team at Alibaba — for the Qwen3.5-9B base and the Qwen3.5 vision-language architecture.
• The SWE-bench authors — Carlos E. Jimenez, John Yang, Alexander Wettig, Shunyu Yao, Kexin Pei, Ofir Press, Karthik Narasimhan, and the OpenAI team behind the Verified split — for the benchmark quoted above: swebench.com.
• Prince Canuma and the mlx-vlm project, and the Apple MLX team, for the tooling that makes vision-language models run natively on Apple Silicon.

The conversion to bf16 MLX is the only work done here. All model quality is theirs.

---

Part of Claude Code Local

This model is one of the fighters in Claude Code Local (3.2k★), which runs Claude Code 100% on-device on Apple Silicon through an MLX-native Anthropic-API server. Not sure which local model to run as an agent? Check the Agent-12 local agent leaderboard: real agent tasks, judged by the filesystem, same hardware for every row.

Built by Matt Macosko in Humboldt, CA. Open to work on local-AI and Apple Silicon inference: matt@ineedhemp.com.