[whatmodel.sh](../../README.md) / MiniCPM5 2B
# MiniCPM5 2B
OpenBMB · released 2026-09 · 2.52B params · apache-2.0
**score 63.1** ±1.0 · #170 of 732 · from 20 benchmark results across 1 source · † no independently run results verified
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks MiniCPM5 2B among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁▃▆█▄●▅▄▄▂▁ #139 of 313 21% 5
coding ▁▁▁▃▅▇█●▅▄▃▁ #162 of 469 33% 5
chat & writing ▁▁▁▃▆▄▆█●▇▇▃ #179 of 402* 0.3% 1
math ▁▁▃▅█▆▄●▄▂▁▁ #144 of 431 29% 4
web search ▅·▅▅█·▃●▃·▅▃ #5 of 15* 4.6% 1
agent tasks · #139 of 313 · 21% of its score
Beats [o3](o3.md), [Gemini 2.5 Pro](gemini-2-5-pro.md), [Gemini 3.1 Flash-Lite](gemini-3-1-flash-lite.md) here, though they rank higher overall. Trails [SmolLM3 3B](smollm3-3b.md), [Jamba 1.5 Large](jamba-1-5-large.md), [GPT-4o (2024-05-13)](gpt-4o-2024-05-13.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
SWE-bench Verified (maker) 46.4% ▄··▄▄█··▄·▄● beats 8/8 very hard · avg 11% 5.6% sharp · at its level · noisy
SWE-bench Pro (split unspecified)… 14.4% █▂···▂···●·▂ beats 7/8 very hard · avg 3.9% 5.2% sharp · noisy
BrowseComp-ZH (maker) 43.5% ▄▄·▄▄▄·█···● beats 8/8 very hard · avg 13% 4.6% at its level · noisy
BFCL V4 (December 2025 archive) (… 66.6% ▄·█·▄▄▄▄·▄·● beats 8/8 medium · avg 43% 4.0% at its level
τ²-bench Telecom (maker) 97.1% ▁·▅█▆▅▂▃▃●▃▁ beats 79/89 medium · avg 35% 1.8% sharp · noisy · small run
coding · #162 of 469 · 33% of its score
Beats [Qwen3-Coder-Next (80B-A3B)](qwen3-coder-next.md), [Gemini 2.5 Pro](gemini-2-5-pro.md), [Gemini 3.1 Flash-Lite](gemini-3-1-flash-lite.md) here, though they rank higher overall. Trails [QwQ-32B-Preview](qwq-32b-preview.md), [Ministral 3 8B Reasoning 2512](ministral-3-8b-reasoning.md), [Phi-4-reasoning 14B](phi-4-reasoning.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench Pro 2025Q2 (Medium)… 17.5% █···▂··▄▂··● beats 8/8 very hard · avg 2.0% 10% sharp · noisy
LiveCodeBench v6 (window unspecif… 69.1% ▃····█·▃▅▃·● beats 8/8 medium · avg 38% 7.0% at its level
SWE-bench Verified (maker) 46.4% ▄··▄▄█··▄·▄● beats 8/8 very hard · avg 11% 5.6% sharp · at its level · noisy
LiveCodeBench Pro 2025Q2 (Easy) (… 68.0% ▄···█▄··▄█▄● beats 8/8 hard · avg 33% 5.6% at its level
SWE-bench Pro (split unspecified)… 14.4% █▂···▂···●·▂ beats 7/8 very hard · avg 3.9% 5.2% sharp · noisy
chat & writing · #179 of 402 · 0.3% of its score
Trails [LFM2.5 1.2B Instruct](lfm2-5-1-2b-instruct.md), [LFM2.5 1.2B Thinking](lfm2-5-1-2b-thinking.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
IFEval (aggregation unspecified) … 86.7% ▄▄····▄··●██ beats 3/8 easy · avg 76% 0.3% noisy
math · #144 of 431 · 29% of its score
Beats [Claude Sonnet 4.6](claude-sonnet-4-6.md), [MiMo-V2.5 (310B-A15B)](mimo-v2-5.md), [MiniMax-M2.7](minimax-m2-7.md) here, though they rank higher overall. Trails [Olmo 3 32B Think](olmo-3-32b-think.md), [Nemotron-Cascade-8B](nemotron-cascade-8b.md), [Olmo 3.1 32B Think](olmo-3-1-32b-think.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HMMT February 2026 (maker) 63.8% ▅··▃▃·▃··▃·● beats 7/8 hard · avg 30% 13% sharp · at its level
AIME 2025 (maker) 86.5% █▄▄▄▄····█·● beats 8/8 medium · avg 42% 7.8% sharp
AIME 2026 (maker) 86.5% ▄·▄█·▄▄···█● beats 8/8 medium · avg 48% 6.2% sharp
MATH-500 (maker) 94.6% ██▄▄●··▄···▄ beats 6/8 very easy · avg 89% 2.5% maxed out
web search · #5 of 15 · 4.6% of its score
Trails [Spark-X2.5 1.7B](spark-x2-5-1-7b.md), [Spark-X2.6 1.7B](spark-x2-6-1-7b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BrowseComp-ZH (maker) 43.5% ▄▄·▄▄▄·█···● beats 8/8 very hard · avg 13% 4.6% at its level · noisy
other tests (7)
test result vs the others in its run difficulty counts why
OJBench (maker) 32.5% ▃····▃▅··█▃● beats 8/8 very hard · avg 12% 6.0% at its level
LongBench v2 (maker) 43.7% ▃······▅▅▃·● beats 7/8 very hard · avg 32% 5.4% at its level
MMLU-Pro (maker) 70.8% ▂··▂█·▂●···▂ beats 7/8 medium · avg 59% 4.4% typical
SuperGPQA (maker) 40.8% ▃·▃·▃▅●····▃ beats 7/8 hard · avg 32% 3.8% at its level
MMLU-Redux (maker) 84.7% ▂···█··▂●··▂ beats 7/8 easy · avg 76% 3.4% typical
IFBench (aggregation unspecified)… 66.3% ▅····▃▃·█·●▃ beats 7/8 medium · avg 44% 2.3% at its level
Multi-IF (aggregation unspecified… 71.8% ▄▄···▄·▄·▄●█ beats 5/8 medium · avg 60% 1.4% typical
source results share of the score
openbmb/MiniCPM5-2B model card (submitt… 20 100% ▇▇▇▇▇▇▇▇▇▇▇▇
prior: pull toward the middle 0% ▇
weighted average = 63.1
Licences: openbmb/MiniCPM5-2B model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run MiniCPM5 2B (Q4_K_M, needs ~1.56 GB, from [huggingface.co/openbmb](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF/tree/2079a22f3beaa4e306449978533478fe0522f4b3))
```sh
llama cli -hf openbmb/MiniCPM5-2B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf openbmb/MiniCPM5-2B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~2.68 GB) `llama cli -hf openbmb/MiniCPM5-2B-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run MiniCPM5 2B (Q4_K_M, needs ~1.56 GB, from [huggingface.co/openbmb](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF/tree/2079a22f3beaa4e306449978533478fe0522f4b3))
```sh
lms get https://huggingface.co/openbmb/MiniCPM5-2B-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~2.68 GB) `lms get https://huggingface.co/openbmb/MiniCPM5-2B-GGUF@Q8_0 -y` · 4bit (~1.4 GB, Mac only) `lms get https://huggingface.co/mlx-community/MiniCPM5-2B-mlx-4Bit -y` · 8bit (~2.7 GB, Mac only) `lms get https://huggingface.co/mlx-community/MiniCPM5-2B-8bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run MiniCPM5 2B (Q4_K_M, needs ~1.56 GB, from [huggingface.co/openbmb](https://huggingface.co/openbmb/MiniCPM5-2B-GGUF/tree/2079a22f3beaa4e306449978533478fe0522f4b3))
```sh
ollama run hf.co/openbmb/MiniCPM5-2B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~2.68 GB) `ollama run hf.co/openbmb/MiniCPM5-2B-GGUF:Q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run MiniCPM5 2B (4bit, needs ~1.4 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/MiniCPM5-2B-mlx-4Bit/tree/86ad5182c5a8532a4681ab972011d34c3b89477a))
```sh
mlx_lm.chat --model mlx-community/MiniCPM5-2B-mlx-4Bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/MiniCPM5-2B-mlx-4Bit
```
Other versions: 8bit (~2.7 GB, Mac only) `mlx_lm.chat --model mlx-community/MiniCPM5-2B-8bit`
Model repository and documentation: [huggingface.co/openbmb](https://huggingface.co/openbmb/MiniCPM5-2B)
[interactive version](https://whatmodel.sh/model/minicpm5-2b) · [JSON](https://whatmodel.sh/api?format=json&all=1)