[whatmodel.sh](../../README.md) / Molmo 2 8B
# Molmo 2 8B
Ai2 · released 2025-12 · 8.66B params · apache-2.0
**score 55.5** ±1.7 · #269 of 732 · from 15 benchmark results across 2 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Molmo 2 8B among every model on that task or test, worst to best.
task vs the others rank counts tests
chat & writing ▁▁▁▃▆▄▆●▆▇▇▃ #203 of 402* 30% 1
math ▁▁▃▅●▆▄█▄▂▁▁ #316 of 431* 8.3% 1
vision & documents ▁▁▂▂▁●▄▂▆▇█▄ #115 of 153 69% 14
chat & writing · #203 of 402 · 30% of its score
Beats [Gemma 2 9B IT SimPO](gemma-2-9b-simpo.md), [Granite 4.2 8B](granite-4-2-8b.md), [Mistral Small 4 (119B-A6.5B)](mistral-small-4.md) here, though they rank higher overall. 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
LMArena (text) 1321.4 ▁▁▁▂▃▃▃●▆▇█▃ beats 154/385 avg 1260 30% blunt
math · #316 of 431 · 8.3% of its score
Trails [DeepSeek R1 Distill Qwen 1.5B](deepseek-r1-distill-qwen-1-5b.md), [Tulu 2 DPO 7B](tulu-2-dpo-7b.md), [Gemma 7B (base)](gemma-7b-base.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
MathVista (testmini) (maker) 58.9% ▁··▁▄▄●▇▄▅█▄ beats 14/48 medium · avg 59% 8.3% at its level · small run
vision & documents · #115 of 153 · 69% of its score
Beats [DeepSeek V4 Flash Vision (Exp)](deepseek-v4-flash-vision-exp.md), [GPT-4 Turbo (2024-04-09)](gpt-4-turbo-2024-04-09.md) here, though they rank higher overall. Trails [Gemma 3 27B](gemma-3-27b.md), [Mistral Small 3.1 24B (2503)](mistral-small-3-1.md), [Llama 4 Scout (109B-A17B)](llama-4-scout.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
Arena vision (style controlled) 1111.9 ▁▁▂▂▂▂●▄▅▅█▇ beats 32/152 avg 1061 20% blunt
MMMU (val) (maker) 53.0% ▁····▁▁▃●▆▆▅ beats 14/53 medium · avg 53% 9.2% at its level · small run
MathVista (testmini) (maker) 58.9% ▁··▁▄▄●▇▄▅█▄ beats 14/48 medium · avg 59% 8.3% at its level · small run
InfographicVQA (maker) 80.1% ▃·▃▅█▃●▃▃▃▃▃ beats 8/13 easy · avg 78% 6.5% small run
CountBench (maker) 93.7% ▂▂▃▂·▂▂▃●·▅█ beats 9/18 very easy · avg 91% 5.2% maxed out · small run
VQAv2 (val) (maker) 87.0% ▃▅▃█·▃▅·▃·▃● beats 13/13 easy · avg 82% 3.2% blunt · small run
ChartQA (maker) 86.0% ▁··▁▁▁▁▁▂▅●█ beats 26/44 easy · avg 83% 2.7% blunt · small run
DocVQA (maker) 93.2% ▁··▃▁▁▁●█▇▄▅ beats 8/28 very easy · avg 93% 2.7% blunt · maxed out · small run
Molmo2-VideoCount accuracy (val) … 35.5% ▃··▃·▃····●▃ beats 3/6 hard · avg 22% 2.6% blunt · at its level
RealWorldQA (maker) 77.6% ▁▃▁▇▃▇▇●█▁▅▅ beats 23/39 easy · avg 73% 2.5% blunt · small run
Ai2 Molmo2 human preference Elo (… 1057.0 ▄··▄··●··▄█▄ beats 2/6 avg 968 2.3% blunt
AI2D test (maker) 95.8% ▁▁▅▇▃▇·▄·█·● beats 27/28 easy · avg 87% 1.3% blunt · small run
Molmo2-CapTest dense video captio… 43.2% █···█·█●██·█ beats 3/6 hard · avg 31% 1.2% blunt · at its level
TextVQA (maker) 85.7% ▁▁▁▄▅▁▃·▃▃▃● beats 24/24 easy · avg 79% 1.2% blunt · small run
source results share of the score
LMArena leaderboard dataset 2 50% ▇▇▇▇▇▇
Molmo2: Open Weights and Data for Visio… 13 49% ▇▇▇▇▇▇
prior: pull toward the middle 1% ▇
weighted average = 55.5
Licences: LMArena leaderboard dataset ([CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)) · Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding ([CC BY 4.0](http://creativecommons.org/licenses/by/4.0/))
## 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 Molmo 2 8B (Q4_K_M, needs ~5 GB, from [huggingface.co/mu2solutions](https://huggingface.co/mu2solutions/Molmo2-8B-GGUF/tree/61c7e7dd2199647c0f6a69b2e14413aeb595fcc8))
```sh
llama cli -hf mu2solutions/Molmo2-8B-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf mu2solutions/Molmo2-8B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~8.7 GB) `llama cli -hf mu2solutions/Molmo2-8B-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 Molmo 2 8B (Q4_K_M, needs ~5 GB, from [huggingface.co/mu2solutions](https://huggingface.co/mu2solutions/Molmo2-8B-GGUF/tree/61c7e7dd2199647c0f6a69b2e14413aeb595fcc8))
```sh
lms get https://huggingface.co/mu2solutions/Molmo2-8B-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 (~8.7 GB) `lms get https://huggingface.co/mu2solutions/Molmo2-8B-GGUF@Q8_0 -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 Molmo 2 8B (Q4_K_M, needs ~5 GB, from [huggingface.co/mu2solutions](https://huggingface.co/mu2solutions/Molmo2-8B-GGUF/tree/61c7e7dd2199647c0f6a69b2e14413aeb595fcc8))
```sh
ollama run hf.co/mu2solutions/Molmo2-8B-GGUF:Q4_K_M
```
Other versions: Q8_0 (~8.7 GB) `ollama run hf.co/mu2solutions/Molmo2-8B-GGUF:Q8_0`
### Other ways to run it
1. Run it as published (fp32, needs ~34.65 GB, NVIDIA only, from [github.com/vllm-project](https://github.com/vllm-project/vllm/blob/64cb68384241f046d07ba3797e1b14dbc82a0b2d/docs/models/supported_models.md))
```sh
vllm serve allenai/Molmo2-8B --trust-remote-code
```
Model repository and documentation: [huggingface.co/allenai](https://huggingface.co/allenai/Molmo2-8B)
[interactive version](https://whatmodel.sh/model/molmo-2-8b) · [JSON](https://whatmodel.sh/api?format=json&all=1)