whatmodel.sh / Mistral 7B Instruct v0.3

Mistral 7B Instruct v0.3

+ contribute

Mistral 7B Instruct v0.3 is an open-weights language model from Mistral, released May 2024. Its published size is 7.25B parameters and a 32k-token context window. Licence: apache-2.0.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2024-05-27). Exact checkpoint Mistral-7B-Instruct-v0.3. The API alias 'open-mistral-7b' is not merged (Mistral lists it as versions 0.2 and 0.3). 7,248,023,552 parameters (safetensors metadata).

39.1±1.1#542 of 719 LLMsfrom 15 benchmark results across 10 sources

$ how it compares

The LLMs ranked either side of Mistral 7B Instruct v0.3, out of 828 with a score, then more from Mistral.

#	model	score	where	RAM	released
632	Granite 3.3 2B Instruct IBM	39±1	local	~3GB	2025-04
633	Phi-3-mini-128k-instruct (Apr 2024) Microsoft	39±2	local	~9.5GB	2024-04
634	Mistral 7B Instruct v0.3 Mistral	39±1	local	~5.8GB	2024-05
635	Vicuna 33B v1.3 LMSYS	39±1	local	~23GB	2023-06
636	DeepSeek-Coder 6.7B Base DeepSeek	39±1	local	~5.8GB	2023-10
	more from Mistral
79	Mistral Large 4 (public preview) Mistral	71±2	api $2.06/M		2026-10
174	Mistral Medium 3.5 (128B) Mistral	65±1	api $3.00/M +local	~42GB	2026-04
180	Mistral Medium 3.1 (2508) Mistral	64±2	comparison only		2025-08

see all LLMs · find the ones that fit your machine

$ what it's good at

coding█··········█ #2 of 2 * · 8.9% of its score · 2 tests

test	result	vs the others in its run	difficulty	counts	why
HumanEval lab	38.4% ✎	▁▁▁▁▁▃▅▅█▃▂▁ beats 51/190	easy · avg 83%	▇▇ 4.9%	sharp · at its level · small run
MBPP lab	50.2% ✎	▁▁▁▄▅█▄▅▃▁▂▁ beats 33/81	easy · avg 76%	▇▇ 4.0%	at its level · small run

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #311 of 393 * · 28% of its score · 2 tests

Beats Gemma 4 E4B, OpenThinker3 7B, Gemma 4 E2B here, though they rank higher overall. Trails LFM2 350M, Gemma 3 1B, Flan-T5 XXL here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
AlpacaEval 2 length-controlled	20.6% ✎	▁▂▂▃▆▆█▆▄▄▁▁ beats 54/122	medium · avg 39%	▇▇▇▇▇▇▇▇▇▇ 25%	typical
Arena-Hard (version/judge unspecified) lab	44.3% ✎	▃▃▂▅▃▆▅█▅▅▂▃ beats 6/30	easy · avg 70%	▇ 3.5%	at its level · small run

math▁▃▃▅▆▅▄█▄▁▁▁ #381 of 398 · 8.4% of its score · 3 tests

Beats Command R (03-2024), Molmo 7B-D (0924) here, though they rank higher overall. Trails Granite 3.1 1B-A400M Instruct, Gemma 2B (base), Llama 3.2 1B Instruct here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
MATH level 5	3.6% ✎	▃▂▅▆▄█▆▃▅▂▃▄ beats 1/92	hard · avg 32%	▇▇ 5.2%	sharp · too hard for it
MATH (subset unspecified) lab	13.2% ✎	▃▄▄▃█▇▅▆▄▃▄▁ beats 2/80	easy · avg 66%	▇ 2.8%	at its level · noisy
OTIS mock AIME 2024–2025	0.3% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 0/193	very hard · avg 8.5%	▇ 0.4%	sharp · too hard for it · noisy

other tests9 not in a task · 54% of its score

test	result	vs the others in its run	difficulty	counts	why
LiveBench 2024-07-26	21.3% ✎	▁▁▁▂▂▃▂█▆▃▅▃ beats 16/61	medium · avg 41%	▇▇▇▇▇▇▇▇▇▇▇▇ 30%	typical
CPTUB average (SpeakLeash, 0-5 judge scale)	55.3% ✎	▁▂▂▂▃█▃▆▅▄▂▃ beats 22/95	easy · avg 66%	▇▇▇ 7.7%	at its level
Polish EQ-Bench v2 (SpeakLeash)	45.2% ✎	▁··▁▁▂▂▆▅▆█▆ beats 15/70	medium · avg 64%	▇▇ 5.8%	at its level
Open Persian LLM Leaderboard v2 average (21 tasks)	41.0% ✎	▂▄▃▃█▄▃▆▃▅▂▂ beats 14/81	medium · avg 58%	▇▇ 4.4%	blunt · at its level
AraGen-v3 3C3H score (Inception, Claude Sonnet 4 judge)	5.1% ✎	▁▁·▁▃▂▄█▄▅▄▄ beats 3/66	hard · avg 31%	▇ 2.7%	typical
Open PT-LLM Leaderboard average (9 Portuguese tasks)	68.1% ✎	▃▁▁▁▁▁▁▁▂█▅▁ beats 144/276	medium · avg 73%	▇ 2.2%	at its level · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)	90.5% ✎	▁▁▂▃▄▄█▆▃▂▂▁ beats 26/140	very easy · avg 94%	▇ 1.0%	blunt · noisy
Chess Puzzles (Epoch)	0.0% ✎	█▁▁▁▁▂▁▄▂▂▁▁ beats 6/135	very hard · avg 5.9%	▇ 0.1%	sharp · too hard for it · noisy

1 more test

test	result	vs the others in its run	difficulty	counts	why
GPQA Diamond	15.2% ✎	▁▁▁▁▂▄▅▅███▄ beats 0/310	hard · avg 40%	—	failed run: Below the 25% guess rate, while Epoch's Llama-3.1-8B-Instruct gets 0.27 and Llama-3-8B 0.26. Probably unparsed answers. Epoch's open-mistral-7b also scores 0.132.

$ where to run it

llama.cpp · recommended

version

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 ↗
Run Mistral 7B Instruct v0.3 · needs ~4.4 GB · from huggingface.co/lmstudio-community
chat in the terminal
llama cli -hf lmstudio-community/Mistral-7B-Instruct-v0.3-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf lmstudio-community/Mistral-7B-Instruct-v0.3-GGUF:Q4_K_M

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Mistral 7B Instruct v0.3 · needs ~4.4 GB · from huggingface.co/lmstudio-community
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/lmstudio-community/Mistral-7B-Instruct-v0.3-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Mistral 7B Instruct v0.3 · needs ~4.85 GB · from ollama.com
chat in the terminal
ollama run mistral:7b-instruct-v0.3-q4_K_M

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Mistral 7B Instruct v0.3 · needs ~4.1 GB, Mac only · from huggingface.co/mlx-community
chat in the terminal
mlx_lm.chat --model mlx-community/Mistral-7B-Instruct-v0.3-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/Mistral-7B-Instruct-v0.3-4bit

Weights and documentation: huggingface.co/mistralai ↗

$ how the score was worked out · LiveBench model judgments (HF… 30%, AlpacaEval 1.0 and 2.0 leader… 25%, Mistral AI vendor-reported be… 15% and 7 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

source	results	share of the score	licence
LiveBench model judgments (HF livebench/model_judgment, 2024-10-22 revision) with the LiveBench question datasets	1	▇▇▇▇▇▇▇▇▇ 30%	Apache-2.0 (stated in the LiveBench datasheet)
AlpacaEval 1.0 and 2.0 leaderboards (tatsu-lab/alpaca_eval repository)	1	▇▇▇▇▇▇▇ 25%	Apache-2.0
Mistral AI vendor-reported benchmark scores for its own models (Hugging Face model cards) · lab-reported	4	▇▇▇▇▇ 15%	Factual vendor-reported score (not licensed content)
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)	1	▇▇ 8%	MIT
SpeakLeash Polish EQ-Bench leaderboard	1	▇▇ 6%	Apache-2.0
Epoch AI — internal evaluations	3	▇▇ 6%	CC BY 4.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)	1	▇ 4%	Apache-2.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)	1	▇ 3%	Apache-2.0
Open Portuguese LLM Leaderboard (archived)	1	▇ 2%	Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇ 1%	Apache-2.0
prior · pull toward the middle		▇ 1%
weighted average		= 39.1

not counted (1)

Epoch AI — internal evaluations

GPQA Diamond [correct]

15.2%

failed run: Below the 25% guess rate, while Epoch's Llama-3.1-8B-Instruct gets 0.27 and Llama-3-8B 0.26. Probably unparsed answers. Epoch's open-mistral-7b also scores 0.132.

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-10-11 02:20 UTC · sources and licences · json