mlx-community/Mellum2-12B-A2.5B-Thinking-4bit

Mellum2 12B A2.5B Thinking - 4-bit MLX

This is a 4-bit affine MLX quantization of JetBrains/Mellum2-12B-A2.5B-Thinking.

Mellum2 Thinking is a reasoning-augmented Mixture-of-Experts assistant model. It has 64 experts with 8 active per token, a 131,072-token context window, and emits reasoning in <think>...</think> blocks before the final answer.

Conversion details

• —Source: JetBrains/Mellum2-12B-A2.5B-Thinking
• —Format: MLX safetensors
• —Quantization: affine, 4 bits, group size 64
• —License: Apache-2.0
• —EOS token: <|im_end|> (token ID 28)

The upstream config.json and generation_config.json identify token ID 0 as the EOS token, while the tokenizer identifies <|im_end|> (ID 28) as EOS. This conversion uses token ID 28 so MLX generation stops at the end of the assistant turn.

Usage

pip install -U mlx-lm

mlx_lm.chat \
  --model mlx-community/Mellum2-12B-A2.5B-Thinking-4bit \
  --max-tokens 8192 \
  --temp 0.6 \
  --top-p 0.95

Benchmarks

Benchmarked with oMLX 0.5.7 and MLX-LM 0.31.3 on an Apple M5 Max MacBook Pro with 128 GB unified memory, using the built-in Python-code workload and 128 generated tokens. Results vary with hardware, runtime versions, context length, and generation settings.

At a 4,096-token prompt, batch generation achieved 193.9, 183.4, and 233.6 aggregate generation tok/s at 2, 4, and 8 concurrent requests, respectively, compared with 95.2 tok/s for one request.

Model provenance

For the original model card, training details, benchmark results, and usage guidance, see the upstream JetBrains checkpoint.