tinyanalysis

Leaderboard / Mistral AI

Ministral 3 8B

Mistral AI · Ministral 3 · released 2025-12-02 · data checked 2026-11-01

Ministral 3 8B is a 8B-parameter dense model from Mistral AI (Europe), released 2025-12-02 under the Apache-2.0 license. It scores 55.2 on the Small LM Index, #7 of 26 small models. It needs about 6.7 GB of memory at 4-bit with an 8K-token context and costs $0.15 per million tokens through the cheapest API provider.

Small LM Index
55.2
#7 of 26 scored small models
Likely 50.5–59.7 · from 3 of 6 core benchmarks
Parameters
8B
Dense
Price per 1M tokens
$0.15
$0.15 in / $0.15 out · blended 3:1
Context window
256K
tokens

Benchmarks

Each bar shows this model's score. The tick marks the best small model on that benchmark. Sources: 3 reported by Mistral AI. How we label sources.

GPQA Diamond Lab-reported Knowledge66.8
Class best: Qwen3.5 9B 81.7
AIME Lab-reported Math · AIME 2025 (Reasoning variant)78.7
Class best: Apriel 1.6 15B Thinker 88.0
LiveCodeBench Lab-reported Coding61.6
Class best: Granite 4.2 8B 73.2

Not reported: estimated

What the index model expects from this model's other scores. Not a measurement.

MMLU-Pro≈ 76
Humanity's Last Exam≈ 8
SWE-bench Verified≈ 38

Specifications

Lab
Mistral AI (FR · Europe)
Architecture
Dense transformer
Total parameters
8B
Active parameters
All (dense)
Size class
Small
Context window
256K tokens
Input modalities
text, image
Reasoning
Hybrid (thinking on/off)
License
Apache-2.0
Openness
Open weights
Memory to run
~6.7 GB at 4-bit, 8K context (more sizes)
Hugging Face
16.3K downloads (30 days) · 120 likes
API providers
1 on OpenRouter
API price
$0.15 in / $0.15 out per 1M (OpenRouter, 2026-11-03)
Weights on Hugging Face ↗Try on OpenRouter ↗

Compare with

Ranked in

Sources

Something wrong here? Report it.

Run it locally

Uses mistralai/Ministral-3-8B-Reasoning-2512-GGUF (Q4_K_M). Swap the tag for another size: BF16, Q4_K_M, Q5_K_M, Q8_0.

Ollama
ollama run hf.co/mistralai/Ministral-3-8B-Reasoning-2512-GGUF:Q4_K_M
llama.cpp
llama-server -hf mistralai/Ministral-3-8B-Reasoning-2512-GGUF:Q4_K_M
Then open http://localhost:8080 for a chat window.
vLLM (GPU server)
vllm serve mistralai/Ministral-3-8B-Reasoning-2512
Full-precision weights, for a server GPU.

Open in LM Studio

Memory needed: weights, context cache and runtime buffers.

Quantization4K context32K context256K context
16-bit F1617 GB21 GB52 GB
8-bit Q8_09.7 GB14 GB45 GB
4-bit Q4_K_M6.1 GB10 GB41 GB

Check it against your GPU, Mac or phone →