Leaderboard / Alibaba (MNN)

MNN-Pocket 0.3B

Alibaba (MNN) · MNN-Pocket · released 2026-09-23 · data checked 2026-11-13

MNN-Pocket 0.3B is a 360M-parameter dense model from Alibaba (MNN) (China), released 2026-09-23 under the Apache-2.0 license. It doesn't have enough reported benchmarks for an index score yet. It needs about 0.8 GB of memory at 4-bit with an 8K-token context.

On-device model co-designed with the MNN inference engine: the INT4 build is 173 MB and decodes at up to 466 tokens/s on an iPad Pro M5. Chinese and English.

Small LM Index
—
Needs 2 core benchmarks (has 0)
Parameters
360M
Dense
Price per 1M tokens
—
No hosted API on OpenRouter; self-host
Context window
32K
tokens

Benchmarks

Each bar shows this model's score. The tick marks the best tiny model on that benchmark. Sources: 3 reported by Alibaba (MNN). How we label sources.

MMLU Lab-reported Knowledge · MMLU-Redux44.0
Class best: MiniCPM5 2B 84.7
IFEval Lab-reported Instruction following73.2
Class best: LFM3 1.7B 96.7
MATH-500 Lab-reported Math35.4
Class best: MiniCPM5 2B 94.6

Specifications

Lab
Alibaba (MNN) (CN · China)
Architecture
Dense transformer
Total parameters
360M
Active parameters
All (dense)
Size class
Tiny
Context window
32K tokens
Input modalities
text
Reasoning
Instruct (no thinking)
License
Apache-2.0
Openness
Open weights
Memory to run
~0.8 GB at 4-bit, 8K context (more sizes)
Hugging Face
5.8K downloads (30 days) · 366 likes

Compare with

Sources

Something wrong here? Report it.

Run it locally

Uses taobao-mnn/MNN-Pocket-0.3B-GGUF (Q8_0). Swap the tag for another size: F16, Q4_1, Q8_0.

Ollama
ollama run hf.co/taobao-mnn/MNN-Pocket-0.3B-GGUF:Q8_0
llama.cpp
llama-server -hf taobao-mnn/MNN-Pocket-0.3B-GGUF:Q8_0
Then open http://localhost:8080 for a chat window.
vLLM (GPU server)
vllm serve taobao-mnn/MNN-Pocket-0.3B
Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization4K context32K context
16-bit F161.3 GB1.7 GB
8-bit Q8_00.9 GB1.3 GB
4-bit Q4_K_M0.8 GB1.2 GB

Check it against your GPU, Mac or phone →