---
title: Mellum2-12B-A2.5B-Thinking-bf16
canonical_url: "https://www.modelscope.cn/models/mlx-community/Mellum2-12B-A2.5B-Thinking-bf16"
md_url: "https://www.modelscope.cn/models/mlx-community/Mellum2-12B-A2.5B-Thinking-bf16.md"
repository: mlx-community/Mellum2-12B-A2.5B-Thinking-bf16
last_updated: 2026-08-11
license: apache-2.0
pipeline_tag: text-generation
tasks:
  - text-generation
model_type:
  - mellum
architectures:
  - MellumForCausalLM
base_model:
  - JetBrains/Mellum2-12B-A2.5B-Thinking
base_model_relation: quantized
parameters: 12.1B
tensor_type:
  - BF16
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 20
stars: 0
tags:
  - mlx
  - mellum
  - moe
  - thinking
  - bf16
---

# Mellum2-12B-A2.5B-Thinking-bf16

> Mellum2-12B-A2.5B-Thinking-bf16 - mlx-community 在 ModelScope 开源的模型。Mellum2 12B A2.5B Thinking - BF16 MLX

mlx-community/Mellum2-12B-A2.5B-Thinking-bf16 是 ModelScope 魔搭社区上的 12.1B 参数text-generation模型，采用 apache-2.0 许可，基于 JetBrains/Mellum2-12B-A2.5B-Thinking 构建。

- **Repository**: mlx-community/Mellum2-12B-A2.5B-Thinking-bf16
- **License**: apache-2.0
- **Tasks**: text-generation
- **Parameters**: 12.1B
- **Base model**: JetBrains/Mellum2-12B-A2.5B-Thinking
- **Tags**: mlx, mellum, moe, thinking, bf16
- **Downloads**: 20
- **Stars**: 0
- **Last updated**: 2026-08-11

Source: https://www.modelscope.cn/models/mlx-community/Mellum2-12B-A2.5B-Thinking-bf16

---

# Mellum2 12B A2.5B Thinking - BF16 MLX

This is an MLX conversion of
[JetBrains/Mellum2-12B-A2.5B-Thinking](https://huggingface.co/JetBrains/Mellum2-12B-A2.5B-Thinking),
stored in bfloat16 without weight quantization.

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
- Precision: bfloat16
- Quantization: none
- 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

Install MLX-LM:

```bash
pip install -U mlx-lm
```

Chat with the model:

```bash
mlx_lm.chat \
  --model mlx-community/Mellum2-12B-A2.5B-Thinking-bf16 \
  --max-tokens 8192 \
  --temp 0.6 \
  --top-p 0.95
```

Or generate a single response:

```bash
mlx_lm.generate \
  --model mlx-community/Mellum2-12B-A2.5B-Thinking-bf16 \
  --prompt "Explain why 1024 is a power of two." \
  --max-tokens 2048 \
  --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
1,024 generated tokens. Results vary with hardware, runtime versions, context
length, and generation settings.

| Prompt tokens | TTFT | Prefill | Generation | End-to-end | Peak memory |
| ---: | ---: | ---: | ---: | ---: | ---: |
| 4,096 | 812.3 ms | 5,042.4 tok/s | 93.8 tok/s | 11.7 s | 23.2 GB |
| 8,192 | 1,707.8 ms | 4,796.9 tok/s | 90.6 tok/s | 13.0 s | 23.4 GB |
| 16,384 | 4,286.3 ms | 3,822.4 tok/s | 79.6 tok/s | 17.2 s | 23.6 GB |
| 32,768 | 10,906.1 ms | 3,004.5 tok/s | 71.6 tok/s | 25.2 s | 23.9 GB |
| 65,536 | 26,813.7 ms | 2,444.1 tok/s | 58.0 tok/s | 44.5 s | 24.6 GB |
| 131,072 | 84,021.3 ms | 1,560.0 tok/s | 35.8 tok/s | 112.6 s | 26.0 GB |

At a 4,096-token prompt, batch generation achieved 60.6, 89.0, and 116.1
aggregate generation tok/s at 2, 4, and 8 concurrent requests, respectively,
compared with 93.8 tok/s for one request.

## Model provenance

For the original model card, training details, benchmark results, and usage
guidance, see the
[upstream JetBrains checkpoint](https://huggingface.co/JetBrains/Mellum2-12B-A2.5B-Thinking).