deskmind/brain-0.8b

DeskMind Brain 0.8b · 得心

得心，应手。 Typed, calibrated step decisions for computer-use agents, running locally on Apple Silicon with MLX. This model is the fast tier (answers every step by default) of the DeskMind Brain router.

• Release: g14-q8 (checkpoint G14, 8-bit MLX, prompt format 3). Earlier releases are on branches g13-q8 and v7b-q8.
• Code and usage: deskmind-ai/brain
• Prompt format: deskmind.json records the format this model was trained with. Keep it next to the weights.

Use

uv run hf download deskmind/brain-0.8b --local-dir models/brain-0.8b
uv run hf download deskmind/brain-4b --local-dir models/brain-4b
uv run deskmind-brain-serve --predictor mlx:models/brain-0.8b --escalate-to mlx:models/brain-4b --two-stage

Results (this 8-bit release, served as shipped)

	DeskMind router (0.8B → 4B)	Jev (TypeSafe AI, cloud reference)
Real macOS desktop, 13 sandbox tasks × 3 runs (bench v23)	92% (35/38 scored), 0 false "done"	87% (33/38 scored), 2 false "done"
Time per step (p50 / p95, M4 Pro)	0.59 s / 4.6 s (local)	0.36 s / 0.44 s (hosted API)

Known limits:

• Slower than cloud models.
• It fails a web-extraction task, which the reference model also fails. When a request is ambiguous, it asks the user before acting.
• It was trained on macOS Finder and TextEdit sandbox tasks plus web and form decisions, and is untested elsewhere.

Training

LoRA distillation on Qwen/Qwen3.5-0.8B (KL to teacher distributions plus cross-entropy to labels), merged and quantized to 8 bits. The data is desktop DAgger in sandboxed tasks labelled by an oracle, plus synthetic and public web and form decision data. It contains no real user data. Details: docs/training.md.

---

中文： 得心（DeskMind）Brain 的快速档（默认回答每一步）。给电脑操作 agent 的每一步做带类型、带把握程度的决策，用 MLX 在 Apple Silicon 本地运行。

• 真机成绩： 13 个沙箱任务各跑 3 轮，通过率 92%，同条件下 Jev 为 87%。
• 速度： 每步中位约 0.6 秒。
• 详见 deskmind-ai/brain。

License

Apache-2.0. Fine-tuned from Qwen/Qwen3.5-0.8B (Copyright Alibaba Cloud, Apache-2.0). The DeskMind name, 得心 and the logo are not covered by this licence.