Mistral AIMistral Large 3VSOpenAIGPT-5 mini

Analysis by:the whichllmmodel Editorial Team|Updated: June 2026

Our Take

We recommend Mistral Large 3 for complex scripting single-file apps or clearly defined functions, or the 1.1x cheaper GPT-5 mini if your budget requires optimizing costs for very high-volume pipelines. While Mistral Large 3 offers a clear reasoning advantage, it carries a moderate price premium. Choose Mistral Large 3 for scripting single-file apps or clearly defined functions, or GPT-5 mini to save on API costs for simple scripts.

▶WHY?

Benchmark Calculations & Evidence:

Coding Evaluation: Mistral Large 3 was evaluated on LiveCodeBench (scoring 34.4%), while GPT-5 mini was evaluated on SWE-bench Pro (scoring 45.7%).

Reasoning Accuracy: Both models were evaluated on the GPQA Diamond benchmark. Mistral Large 3 scored 85.5%, while GPT-5 mini scored 81.6%.

Cost Efficiency: GPT-5 mini pricing ($0.25/M input, $2/M output) is 1.1x cheaper than Mistral Large 3 ($0.5/M input, $1.5/M output).

Was this recommendation helpful?

Model Specs

Mistral Large 3

Open SourceAPI Available

Website 🤗HF

Benchmarks & Scores

Coding (live-code-bench)

34.4%

scripting single-file apps or clearly defined functions

Reasoning (gpqa-diamond)Winner (+3.9%)

85.5%

graduate-level science QA

Cost & Context

Cost (per 1M tokens)

$0.75Input: $0.50 | Output: $1.50

Context Window

262.14k tokens

Model Specs

GPT-5 mini

Website

Benchmarks & Scores

Coding (swe-bench-pro)

45.7%

complex codebases, multi-file repositories, and architectural planning

Reasoning (gpqa-diamond)

81.6%

graduate-level science QA

Cost & Context

Cost (per 1M tokens)1.1x cheaper

$0.69Input: $0.25 | Output: $2.00

Context WindowLarger

400k tokens

Read our data collection methodology

Frequently Asked Questions about Mistral Large 3 vs GPT-5 mini

GPT-5 mini is cheaper than Mistral Large 3. GPT-5 mini has a blended cost of $0.69/1M tokens, which is about 1.1x cheaper than Mistral Large 3 at $0.75/1M tokens.

For coding tasks, Mistral Large 3 scores 34.4% on live-code-bench (scripting single-file apps or clearly defined functions), while GPT-5 mini scores 45.7% on swe-bench-pro (complex codebases, multi-file repositories, and architectural planning).

Related Matchups

Explore similar comparisons for Mistral Large 3 and GPT-5 mini.

Browse More Comparisons

Mistral AIMistral Large 3

Mistral AIMistral Large 3

OpenAIGPT-5.6 Terra

Compare Specs

Do you want to find a model for your constraints?

Use our interactive model finder to filter LLMs by reasoning capability, coding performance, cost, and context length.

Open Model Finder