Ministral 3B vs Ministral 8B Instruct vs Llama-3.1-8B-Instruct

It is close. Our weighted score puts them within 1 points (Ministral 3B 43/100, Ministral 8B Instruct 42/100, Llama-3.1-8B-Instruct 42/100), so choose by what matters most for your work: Ministral 8B Instruct for raw capability and Ministral 3B on price. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%. Capability uses the average of shared benchmarks (GPQA Diamond), because Ministral 8B Instruct has no Capabilities Index score yet.

Ministral 3B is cheaper at $0.10 input / $0.10 output per million tokens (median across 1 API provider). Ministral 8B Instruct costs $0.15 input / $0.15 output per million tokens (median across 1 API provider); Llama-3.1-8B-Instruct costs $0.152 input / $0.167 output per million tokens (median across 9 API providers). At a typical mix of three input tokens to one output token, that is $0.10 per million tokens for Ministral 3B versus $0.15 for Ministral 8B Instruct (1.5× as much) and $0.156 for Llama-3.1-8B-Instruct (1.6× as much).

Not every model here has a Capabilities Index score, so we compare the average of shared benchmarks (GPQA Diamond): Ministral 8B Instruct 27.2%, Llama-3.1-8B-Instruct 27.0% and Ministral 3B 25.3%. On individual benchmarks: GPQA Diamond — Ministral 8B Instruct 27.2%, Llama-3.1-8B-Instruct 27.0%, Ministral 3B 25.3%.

There are no published SWE-bench Verified results for Ministral 3B, Ministral 8B Instruct and Llama-3.1-8B-Instruct yet, so there is no like-for-like coding score. On overall capability, Ministral 8B Instruct leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Ministral 8B Instruct has the largest context window at 131,072 tokens, against 128,000 for Ministral 3B and 128,000 for Llama-3.1-8B-Instruct. Maximum output per response: Ministral 3B up to 8,192, Ministral 8B Instruct up to 8,192, Llama-3.1-8B-Instruct up to 4,096 tokens.

Ministral 3B accepts text; Ministral 8B Instruct accepts text; Llama-3.1-8B-Instruct accepts text. They handle the same number of input types.

Yes, all three publish their weights (Mistral Research License), so you can self-host them.

Ministral 3B is the newest, released Oct 16, 2024. Ministral 8B Instruct came out Oct 16, 2024; Llama-3.1-8B-Instruct came out Jul 23, 2024. Knowledge cutoff: Ministral 3B Mar 2024, Llama-3.1-8B-Instruct Dec 2023.

Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.