Mistral Large 3
Mistral
- Endpoint
- —
- Model parameter
- —
- Canonical ID
- mistral-large-2512
See the model page for a full documentation-style example request and the citation it derives from.
Side-by-side reference
Mistral Large 3 vs Gemini 2.5 Pro
Two-sided verified comparison: Mistral's current Large-tier flagship (v25.12) and Google Gemini 2.5 Pro. Both sides are verified end-to-end from primary vendor documentation. This page does not declare a winner.
Data freshness: last checked 8 d ago · updated 2026-05-21
At a glance
Attributes
| Attribute | Mistral Large 3 | Gemini 2.5 Pro |
|---|---|---|
| Provider | Mistral | |
| Snapshot date | Data not yet verified. | Data not yet verified. |
| Context window (tokens) | 256,000src | 1,048,576src |
| Max output (tokens) | Data not yet verified. | 65,536src |
| Modality | Data not yet verified. | text-in, image-in, audio-in, video-in, text-outsrc |
| Lifecycle | activesrc | activesrc |
This table compares verified attributes only; cells with no verified value display the canonical unverified-data label. WebmasterID Models does not declare a winner.
Reference
First-party pricing references
Per-unit rates published by each model's creator on their own first-party API. Rows are source-backed references, not live quotes. Hosted-provider references (Groq, Together) render separately below where available.
Pricing is volatile and may change after retrieval. Treat prices as source-backed references, not live quotes. Prices are shown as source-backed references only. WebmasterID Models does not rank models or billing providers by price.
Gemini 2.5 Pro
| Unit | Price (USD) |
|---|---|
| 1M input tokens | $1.25src |
| 1M input tokens (>200k context) | $2.50src |
| 1M output tokens | $10.00src |
| 1M output tokens (>200k context) | $15.00src |
| 1M cache write tokens (5m) | $0.13src |
| 1M cache write tokens (>200k context) | $0.25src |
| 1M cache storage / hour | $4.50src |
| 1M batch input tokens | $0.63src |
| 1M batch input tokens (>200k context) | $1.25src |
| 1M batch output tokens | $5.00src |
| 1M batch output tokens (>200k context) | $7.50src |
Capability
Benchmarks
Mistral Large 3
Data not yet verified.Gemini 2.5 Pro
Data not yet verified.Where to use
Use cases
- Long-context document analysis (Gemini 2.5 Pro 1,048,576 tokens; Mistral Large 3 256k tokens)
- Open-weights European serving (Mistral) versus closed-weights US-hosted (Google)
- Multimodal understanding — Gemini publishes audio/image/video/text/PDF inputs explicitly; Mistral's card describes the model as multimodal without enumerating channels
API surface
API usage differences
Endpoint shape, model parameter position, and request body conventions for each provider. Shown for reference, not as a recommendation.
Gemini 2.5 Pro
- Endpoint
- POST https://generativelanguage.googleapis.com/v1beta/models/gemini-2.5-pro:generateContent
- Model parameter
- URL path segment
- Canonical ID
- gemini-2.5-pro
See the model page for a full documentation-style example request and the citation it derives from.
Caveats
Limitations of this comparison
- Mistral Large 3's max output and explicit modality channels are not stated on the vendor spec card — those fields render unverified.
- Gemini publishes prompt-size tier pricing (≤200k vs >200k) and per-hour cache storage; Mistral's pricing surface is base input/output only on the spec card. Direct unit-by-unit normalisation is not meaningful.
- WebmasterID Models does not declare a winner; readers should compare verified facts against their own workload.
Source trail
- Official docsRetrieved 2026-05-21FreshMistral — Large 3 (v25.12) model card
Mistral Large 3 spec card at the new URL pattern (/models/model-cards/<slug>). Canonical pinned snapshot is mistral-large-2512; context window 256k; input $0.5 / output $1.5 per 1M tokens. Max output and explicit modality list still not stated.
- Official docsRetrieved 2026-05-20FreshMistral — Models overview
Mistral models overview — current active models, version stamps, open vs premier classification.
- Official docsRetrieved 2026-05-20FreshMistral — Models table (API strings)
Mistral models table — exact API string for every current model (mistral-large-3, mistral-medium-3-5, etc.).
- Official docsRetrieved 2026-05-20FreshMistral — API reference
Mistral chat completions endpoint, headers, minimal request body, and model parameter format.
- Official docsRetrieved 2026-05-20FreshGoogle AI — Gemini 2.5 Pro model docs
Per-model specification page for gemini-2.5-pro (model code, token limits, supported input types, knowledge cutoff, supported features).
- Official pricingRetrieved 2026-05-20FreshGoogle AI — Gemini pricing
Gemini API pricing reference — standard, batch, flex, and priority tiers with ≤200k vs >200k-token pricing.
- Official docsRetrieved 2026-05-20FreshGoogle AI — Gemini API quickstart
Quickstart with REST endpoint shape, model parameter format, request body, and Python/JS SDK examples.
Continue
Related
Want a new comparison? See docs for how WebmasterID Models verifies entries.