Codestral 2 vs Gemini 3 Pro
A complete head-to-head: pricing, context window, benchmarks, modality coverage, and openness, with a programmatic verdict synthesized from the underlying data.
Updated
Codestral 2 specs · Gemini 3 Pro specs- PriceCodestral 2
Codestral 2 publishes pricing ($0.90 / 1M output tokens) while Gemini 3 Pro does not.
- Context windowGemini 3 Pro
Gemini 3 Pro accepts 1.0M tokens vs 256K, 4.1× the room for long documents and codebases.
- BenchmarksGemini 3 Pro
Gemini 3 Pro leads in 1 of 1 shared benchmarks; the biggest gap is on HumanEval (Python coding), where it scores 95.4 vs 92.0.
- ModalitiesGemini 3 Pro
Gemini 3 Pro supports 4 modalities (text, vision, audio, video) vs 1 for Codestral 2.
- OpennessCodestral 2
Codestral 2 ships open weights (Mistral Non-Production License); Gemini 3 Pro is API-only.
On balance Gemini 3 Pro edges ahead, winning 3 of 5 categories against Codestral 2's 2. Codestral 2 publishes pricing ($0.90 / 1M output tokens) while Gemini 3 Pro does not. Gemini 3 Pro accepts 1.0M tokens vs 256K, 4.1× the room for long documents and codebases.
Gemini 3 Pro leads in 1 of 1 shared benchmarks; the biggest gap is on HumanEval (Python coding), where it scores 95.4 vs 92.0. They differ in modality coverage, Codestral 2 handles text while Gemini 3 Pro handles text, vision, audio, video, which can be the deciding factor before you even look at benchmarks. Codestral 2 ships open weights (Mistral Non-Production License); Gemini 3 Pro is API-only.
Gemini 3 Pro is the newer of the two, released 11 months after Codestral 2, which usually means a more recent knowledge cutoff and updated safety post-training. Codestral 2 is usually picked for code completion and coding llm workloads, while Gemini 3 Pro sees more deployments in long context and rag. If pricing matters more than every last benchmark point, run the numbers in the calculator below before committing.
Side-by-side specs
| Spec | Codestral 2 | Gemini 3 Pro |
|---|---|---|
| Provider | Mistral | |
| Released | Jan 2025 | Dec 2025 |
| Modalities | text | text, vision, audio, video |
| Context window | 256K tokens | 1.0M tokens |
| Max output | , | 65.5K tokens |
| Input · 1M | $0.30 / 1M tokens | Pricing not published |
| Output · 1M | $0.90 / 1M tokens | Pricing not published |
| Knowledge cutoff | , | 2025-01 |
| Open weights | Yes (Mistral Non-Production License) | No |
| API available | Yes | Yes |
Pricing at scale
What you'd actually pay at typical workloads. Numbers come from each model's published per-million-token rates.
- Light usage, 100k in / 50k out per day$2.25 vs ,
- Heavy usage, 1M in / 500k out per day$22.50 vs ,
- RAG workload, 5M in / 200k out per day$50.40 vs ,
Light usage, 100k in / 50k out per day: pricing not directly comparable (one or both models are missing public per-token rates). Heavy usage, 1M in / 500k out per day: pricing not directly comparable (one or both models are missing public per-token rates). RAG workload, 5M in / 200k out per day: pricing not directly comparable (one or both models are missing public per-token rates).
Estimated spend for the listed models at your usage. Numbers are derived from each model's published per-million-token rates.
- Codestral 2$0.075
- Gemini 3 ProPricing unavailable
Benchmarks compared
Only sourced numbers. Where a benchmark is missing for one model we show the available value rather than fabricating the other.
- MMLU,91.8
- HumanEval92.095.4
- GPQA,91.9
Codestral 2 fits when…
- Fast inline completion
- FIM support
- Self-hostable for IP-sensitive teams
- Self-hosting and on-prem requirements, open weights (Mistral Non-Production License).
Gemini 3 Pro fits when…
- Massive 1M-token context window
- State-of-the-art vision and document understanding
- Strong reasoning at competitive price
- Long-context tasks, handles 1.0M tokens vs 256K for Codestral 2.
- Multimodal needs covering vision, audio, video.
Consider Mistral Large 2
Mistral's flagship API model, strong on code and reasoning, EU-friendly hosting.
Frequently asked
Is Codestral 2 or Gemini 3 Pro cheaper?
Per-token pricing isn't published for at least one of these models, check each model's spec page for current rates.Which has the larger context window?
Gemini 3 Pro accepts 1.0M tokens vs 256K for Codestral 2.Is Codestral 2 or Gemini 3 Pro better for coding?
Both Codestral 2 and Gemini 3 Pro are competitive on coding benchmarks. See each model's individual spec page for HumanEval and SWE-bench scores where published. For an opinionated pick, consult our Best LLM for Coding ranking.Are either of these models open source?
Codestral 2 ships open weights (Mistral Non-Production License). Gemini 3 Pro is API-only.When were Codestral 2 and Gemini 3 Pro released?
Codestral 2 was released by Mistral on 2025-01-14. Gemini 3 Pro was released by Google on 2025-12-09.
The week's AI launches, in your inbox.
One short email every Friday, new models, leaks, and quietly-shipped APIs you missed.