whatmodel.sh / Mistral NeMo Instruct (2407)

Mistral NeMo Instruct (2407)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Mistral NeMo Instruct (2407) is an open-weights language model from Mistral, released July 2024. Its published size is 12.25B parameters and a 128k-token context window. Licence: apache-2.0.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2024-07-18). Epoch label is the Mistral API id open-mistral-nemo-2407, which Mistral's legacy-models table lists as 'Mistral Nemo 12B' version 24.07 (deprecated 5/22/2026, retired 7/31/2026). Same release as open weights Mistral-Nemo-Instruct-2407 (12,247,782,400 parameters).

44.4±0.8#449 of 719 LLMsfrom 23 benchmark results across 15 sources

$ how it compares

The LLMs ranked either side of Mistral NeMo Instruct (2407), out of 828 with a score, then more from Mistral.

#modelscorewhereRAMreleased
536Llama 3.1 8B Instruct Meta45±0api $0.22/M +local~6.2GB2024-07
537Command R Plus (04-2024) Cohere45±1local~43GB2024-04
538Mistral NeMo Instruct (2407) Mistral44±1local~8.7GB2024-07
539LLaVA-OneVision Qwen2 72B (ov-sft) LLaVA (lmms-lab)44±3local~155GB2024-08
540Phi-3-mini-4k-instruct (June 2024 update) Microsoft44±1local~3.7GB2024-06
more from Mistral
79Mistral Large 4 (public preview) Mistral71±2api $2.06/M2026-10
174Mistral Medium 3.5 (128B) Mistral65±1api $3.00/M +local~42GB2026-04
180Mistral Medium 3.1 (2508) Mistral64±2comparison only2025-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #256 of 300 · 4.2% of its score · 3 tests

Beats DeepSeek R1 0528 Qwen3 8B, Magistral Small 1.0 (2506), Llama 4 Scout (109B-A17B) here, though they rank higher overall. Trails Granite 4.0 H 1B, Granite 4.0 1B, Qwen2.5 3B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL V4 (December 2025 archive)27.6% ✎▁▁▁▂██▅▅▆▂▂▂ beats 22/77hard · avg 32%▇ 1.9%typical
Galileo Agent Leaderboard v1 (tool selection quality)66.1% ✎▃▃▃▃▆▅▆▅█▂▅▂ beats 1/31easy · avg 79%▇ 1.1%blunt · at its level
BFCL v351.5% ✎▁▁▁▁▁▃▅█▅▄▁▁ beats 40/114medium · avg 52%▇ 1.1%blunt · at its level
coding▁▁▁▃▆██▇▆▄▃▁ #265 of 458 · 3.5% of its score · 4 tests

Beats Qwen3-VL 32B Instruct, Solar Open 100B (102B-A12B), GLM-4.7-Flash here, though they rank higher overall. Trails DeepSeek LLM 67B Chat here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
CodeReviewQA: solution identification (hard), invariant accuracy52.8% ✎▃··▁▂▂▃▂█▃▄▂ beats 28/47medium · avg 54%▇ 1.3%sharp · at its level · noisy
CodeReviewQA: change localisation (hard), invariant accuracy29.0% ✎█▂▁▁▂▂▂▃▅▅▅▇ beats 24/47medium · avg 36%▇ 1.2%sharp · noisy
CodeReviewQA: automated code refinement, exact match rate33.1% ✎▁▁▁▁▁▁▂▂▃█▅▄ beats 31/47hard · avg 33%▇ 1.0%at its level
CodeReviewQA: change type recognition, invariant accuracy75.8% ✎▁▁▁······▁▃█ beats 32/47easy · avg 67%▇ 0.0%blunt · at its level · noisy
math▁▂▃▅▆▅▄█▄▁▁▁ #345 of 398 * · 16% of its score · 2 tests

Beats DBRX Instruct, Command R Plus (04-2024) here, though they rank higher overall. Trails Qwen3 0.6B, Gemma 7B (base), Qwen2.5 1.5B Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
GSM8K (ZeroEval, zero-shot)82.8% ✎▁▄·▁▄█▅▅▅███ beats 15/43very easy · avg 89%▇▇▇▇ 10%sharp
MATH level 510.8% ✎▂▂▅▆▄█▆▃▅▂▃▄ beats 6/92hard · avg 32%▇▇ 6.2%sharp
other tests14 not in a task · 76% of its score
testresultvs the others in its rundifficultycountswhy
MMLU-Redux (ZeroEval, 2,778 questions)66.9% ✎▁·▁▁▄▄▄▅█▄▂▁ beats 9/53easy · avg 77%▇▇▇▇▇▇▇▇▇▇ 26%at its level
LiveBench 2024-07-2629.2% ✎▁▁▁▂▂▃▂█▆▃▅▃ beats 24/61medium · avg 41%▇▇▇▇▇▇▇▇ 19%at its level
WildBench v2 WB-Score (task macro)44.3751 WB-Score ✎▁·▁▁·▃█▄▇▆▅▆ beats 32/52avg 45.199 WB-Score▇▇ 5.8%sharp
ZebraLogic (ZeroEval, 1,000 puzzles)11.8% ✎▁▁▃▆██▁▁·▁▁▁ beats 13/57hard · avg 18%▇▇ 4.3%sharp
EvalPerf correctness (118 tasks, 20240328)57.4% ✎▂·▃▂▂▆▅▆▆▆▃█ beats 1/29easy · avg 76%▇▇ 3.9%at its level
CPTUB average (SpeakLeash, 0-5 judge scale)58.8% ✎▁▂▂▂▃█▃▆▅▄▂▃ beats 27/95easy · avg 66%▇▇ 3.8%at its level
GPQA Diamond29.9% ✎▁▁▁▃▂▄▆▄▅▇█▅ beats 16/208hard · avg 40%▇ 3.5%sharp
Polish EQ-Bench v2 (SpeakLeash)61.8% ✎▁··▁▁▂▂▆▅▆█▆ beats 42/70medium · avg 64%▇ 3.0%at its level
6 more tests
testresultvs the others in its rundifficultycountswhy
Open Persian LLM Leaderboard v2 average (21 tasks)46.7% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 21/81medium · avg 58%▇ 2.2%blunt · at its level
RULER 64k46.8% ✎▃▁·▁·▂▃▄█▂·▁ beats 0/22easy · avg 83%▇ 1.4%typical
Open PT-LLM Leaderboard average (9 Portuguese tasks)73.8% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 210/276medium · avg 73%▇ 1.1%at its level · noisy
EuroEval Life in the UK (English knowledge MC), test split, 5-shot, accuracy84.6% ✎▁▁▁▁▁▁▁▂▄█▄▃ beats 97/146easy · avg 83%▇ 0.7%at its level · noisy
RULER 128k19.0% ✎▄··▂▄▂▆██▄▄▂ beats 0/21easy · avg 68%▇ 0.5%at its level · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)88.8% ✎▁▁▂▃▄▄█▆▃▂▂▁ beats 20/140very easy · avg 94%▇ 0.4%blunt · maxed out · noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run Mistral NeMo Instruct (2407) · needs ~7.5 GB · from huggingface.co/lmstudio-community
    chat in the terminal
    llama cli -hf lmstudio-community/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf lmstudio-community/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run Mistral NeMo Instruct (2407) · needs ~7.5 GB · from huggingface.co/lmstudio-community

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/lmstudio-community/Mistral-Nemo-Instruct-2407-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Mistral NeMo Instruct (2407) · needs ~7.85 GB · from ollama.com
    chat in the terminal
    ollama run mistral-nemo:12b

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run Mistral NeMo Instruct (2407) · needs ~6.9 GB, Mac only · from huggingface.co/mlx-community
    chat in the terminal
    mlx_lm.chat --model mlx-community/Mistral-Nemo-Instruct-2407-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model mlx-community/Mistral-Nemo-Instruct-2407-4bit

Weights and documentation: huggingface.co/mistralai ↗

$ how the score was worked out · WildEval ZeroEval published r… 40%, LiveBench model judgments (HF… 19%, Epoch AI — internal evaluatio… 10% and 12 more

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
WildEval ZeroEval published results3▇▇▇▇▇▇▇▇▇▇▇▇ 40%Apache-2.0
LiveBench model judgments (HF livebench/model_judgment, 2024-10-22 revision) with the LiveBench question datasets1▇▇▇▇▇▇ 19%Apache-2.0 (stated in the LiveBench datasheet)
Epoch AI — internal evaluations2▇▇▇ 10%CC BY 4.0
WildBench v2 leaderboard data (Ai2)1▇▇ 6%Apache-2.0
EvalPlus / EvalPerf published evaluation runs1▇ 4%Apache-2.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇ 4%MIT
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)4▇ 4%CC-BY-4.0
Berkeley Function Calling Leaderboard result archive2▇ 3%Apache-2.0
SpeakLeash Polish EQ-Bench leaderboard1▇ 3%Apache-2.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇ 2%Apache-2.0
NVIDIA RULER repository README results table2▇ 2%Apache-2.0
Galileo Agent Leaderboard (v1 and v2 results)1▇ 1%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 1%Apache-2.0
EuroEval leaderboards (results collection), English generative leaderboard1▇ 1%MIT
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 0%Apache-2.0
prior · pull toward the middle▇ 0%
weighted average= 44.4

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json