[whatmodel.sh](../../README.md) / Olmo 3 7B Instruct
# Olmo 3 7B Instruct
Ai2 · released 2025-11 · 7.3B params · apache-2.0
**score 46.1** ±0.8 · #454 of 732 · from 29 benchmark results across 5 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Olmo 3 7B Instruct among every model on that task or test, worst to best.
task vs the others rank counts tests
agent tasks ▁▁●▆█▄▅▅▄▄▂▁ #301 of 313 3.5% 3
coding ▁▁▁▃▅●█▇▅▄▃▁ #353 of 469 13% 5
chat & writing ▁▁▁▃▆▄▆●▆▇▇▃ #224 of 402 4.6% 2
math ▁▁▃▅█●▄█▄▂▁▁ #201 of 431 19% 4
agent tasks · #301 of 313 · 3.5% of its score
Beats [QwQ-32B-Preview](qwq-32b-preview.md), [DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B)](deepseek-coder-v2-lite-instruct.md) here, though they rank higher overall. Trails [Granite 4.0 350M](granite-4-0-350m.md), [Granite 4.0 H 350M](granite-4-0-h-350m.md), [ChatGLM3-6B](chatglm3-6b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
BFCL V4 (December 2025 archive) 19.8% ▁▁▁●▆▆▅▆█▂▄▃ beats 6/114 medium · avg 36% 1.4% at its level
BFCL (version not stated) (maker) 49.8% ●▆▄▂·█▆·▄··▂ beats 0/17 medium · avg 63% 1.3% blunt · at its level
BFCL v3 41.9% ▁▁▁▁▁●▅█▆▅▄▁ beats 21/171 medium · avg 55% 0.8% blunt · at its level
coding · #353 of 469 · 13% of its score
Beats [LFM2.5 2.6B](lfm2-5-2-6b.md), [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md), [Qwen3-VL 2B Thinking](qwen3-vl-2b-thinking.md) here, though they rank higher overall. Trails [Phi-3-mini-128k-instruct (Apr 2024)](phi-3-mini-128k-instruct.md), [DeepSeek LLM 67B Chat](deepseek-llm-67b-chat.md), [OpenChat 3.5](openchat-3-5.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
LiveCodeBench v6 (window unspecif… 28.2% ▂●▄▄▅██▆▄▇▄▆ beats 2/67 hard · avg 33% 3.8% typical
HumanEval+ (maker) 77.2% ▂·······●·▂█ beats 1/6 easy · avg 78% 3.1% sharp · at its level · noisy
MBPP+ (maker) 60.2% ●▄····▄·▄··█ beats 0/5 medium · avg 61% 2.5% at its level
LiveCodeBench v3 (maker) 29.5% ●▃·····▃·▃·█ beats 0/6 medium · avg 64% 2.5% at its level
MultiPL-E 36.1% ▃··●·▅▅▇▄█▃▄ beats 3/31 medium · avg 54% 1.1% at its level
chat & writing · #224 of 402 · 4.6% 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) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AlpacaEval 2 length-controlled (m… 40.9% █···●·█·█·██ beats 1/5 medium · avg 43% 3.1% sharp · at its level
IFEval (aggregation unspecified) … 85.6% ▄▄·····●▄▄·▄ beats 2/6 easy · avg 75% 1.5% at its level · several runs
math · #201 of 431 · 19% of its score
Beats [MiniMax-M2.1](minimax-m2-1.md), [DeepSeek V3 (Dec '24)](deepseek-v3.md), [Qwen3 235B-A22B (Apr 2025, hybrid)](qwen3-235b-a22b.md) here, though they rank higher overall. Trails [Mixtral 8x7B v0.1 (base)](mixtral-8x7b-v0-1.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
AIME 2025 (maker) 32.5% ▄··●····█▄▄▄ beats 1/6 medium · avg 50% 8.6% sharp · at its level
AIME 2024 (maker) 44.3% ▄·····●··▄██ beats 1/6 medium · avg 58% 7.8% sharp · at its level
MATH (subset unspecified) (maker) 87.3% ██····██·█●█ beats 5/6 easy · avg 70% 1.3% at its level · noisy
OMEGA (math generalisation) (make… 28.9% █···▄·▄·▄··● beats 5/6 hard · avg 19% 1.2% typical
other tests (14)
test result vs the others in its run difficulty counts why
Medmarks-V mean win rate (19 veri… 0.3503 ▁●▃▃▇▄█▃▅▁▁▁ beats 1/30 avg 0.418 11% sharp
BIG-Bench Hard (maker) 71.2% ▄·····●·▄·██ beats 1/6 easy · avg 77% 10% sharp · at its level
MMLU (maker) 69.1% ▄···●··█··▄█ beats 1/6 easy · avg 76% 6.7% at its level
GPQA Diamond 40.9% ▁▁▁▁▂●▅▅███▄ beats 52/322 hard · avg 46% 5.9% sharp
GPQA (subset unspecified) (maker) 40.4% ▃······●·▅·█ beats 1/6 hard · avg 43% 5.6% sharp
MMLU-Redux 71.8% ▁▁●▁▂▃▄▅▇▅█▄ beats 4/63 easy · avg 78% 4.6% at its level
Medmarks-V weighted mean win rate… 35.0% ▂●▅▂▆▃█▅█▅▃▆ beats 1/34 medium · avg 41% 4.5% blunt · at its level
AGIEval English (maker) 64.4% ●····▄·▄▄··█ beats 0/5 easy · avg 76% 3.8% at its level
ZebraLogic (maker) 32.9% ●▄····▄·▄·▄▄ beats 1/6 medium · avg 52% 2.6% at its level
PopQA (maker) 14.1% ●··█···▄···█ beats 0/5 hard · avg 20% 1.6% typical
IFBench (aggregation unspecified)… 32.3% █·●█·██····█ beats 1/5 medium · avg 36% 1.2% at its level
EuroEval HellaSwag (English), tes… 61.3% ▃·▁▁▂▁▂●█▇▄▁ beats 36/100 medium · avg 67% 1.2% sharp · at its level · noisy
GSM-Plus 85.8% ▁··▁▁▁·▃▁●█▁ beats 7/16 easy · avg 82% 0.9% blunt
EuroEval Life in the UK (English … 75.1% ▁▁▁▁▁▁▁▂●█▄▃ beats 49/146 easy · avg 82% 0.9% at its level · noisy
source results share of the score
Olmo 3 (submitted / maker-reported) 14 59% ▇▇▇▇▇▇▇
JetBrains/Mellum2-12B-A2.5B-Instruct mo… 8 19% ▇▇
Medmarks: A Comprehensive Open-Source L… 2 15% ▇▇
allenai/Olmo-3-7B-Instruct model card (… 3 4% ▇
EuroEval leaderboards 2 2% ▇
prior: pull toward the middle 0% ▇
weighted average = 46.1
Licences: Olmo 3 ([CC BY 4.0](../../licenses/report-2512.13961v2.txt)) · JetBrains/Mellum2-12B-A2.5B-Instruct model card ([Apache-2.0](https://huggingface.co/JetBrains/Mellum2-12B-A2.5B-Instruct/blob/273d9c1c6882ac8e1562e5b7e043857d35a787b6/README.md)) · Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks ([CC BY 4.0](../../licenses/paper-2605.01417v1-CC-BY-4.0.txt)) · allenai/Olmo-3-7B-Instruct model card ([Apache-2.0](../../licenses/allenai--Olmo-3-7B-Instruct.txt)) · EuroEval leaderboards ([MIT](https://github.com/EuroEval/leaderboards/blob/1a77f1d2f8ba1ba05ab73287fb090d76bf41b2e0/LICENSE))
## 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 Olmo 3 7B Instruct (Q4_K_M, needs ~4.5 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-GGUF/tree/893e8b40befaf8a27cadd1349c0a9cbbe529cffd))
```sh
llama cli -hf lmstudio-community/Olmo-3-7B-Instruct-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf lmstudio-community/Olmo-3-7B-Instruct-GGUF:Q4_K_M
```
Other versions: Q8_0 (~7.8 GB) `llama cli -hf lmstudio-community/Olmo-3-7B-Instruct-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 Olmo 3 7B Instruct (Q4_K_M, needs ~4.5 GB, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-GGUF/tree/893e8b40befaf8a27cadd1349c0a9cbbe529cffd))
```sh
lms get https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-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 (~7.8 GB) `lms get https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-GGUF@Q8_0 -y` · 4bit (~4.1 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-MLX-4bit -y` · 8bit (~7.8 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-MLX-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 Olmo 3 7B Instruct (Q4_K_M, needs ~4.88 GB, from [ollama.com](https://ollama.com/library/olmo-3/tags))
```sh
ollama run olmo-3:7b-instruct
```
Other versions: Q8_0 (~9.26 GB) `ollama run olmo-3:7b-instruct-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 Olmo 3 7B Instruct (4bit, needs ~4.1 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Olmo-3-7B-Instruct-MLX-4bit/tree/595846233eb3a955ab3e2227ebc51a35db58fef8))
```sh
mlx_lm.chat --model lmstudio-community/Olmo-3-7B-Instruct-MLX-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model lmstudio-community/Olmo-3-7B-Instruct-MLX-4bit
```
Other versions: 8bit (~7.8 GB, Mac only) `mlx_lm.chat --model lmstudio-community/Olmo-3-7B-Instruct-MLX-8bit`
Model repository and documentation: [huggingface.co/allenai](https://huggingface.co/allenai/Olmo-3-7B-Instruct)
[interactive version](https://whatmodel.sh/model/olmo-3-7b-instruct) · [JSON](https://whatmodel.sh/api?format=json&all=1)