Mistral Large 3 API Pricing: Best Quality-to-Price at $0.50/M
Mistral Large 3 offers strong reasoning with a 262K context window at $0.50/M input, $1.50/M output — the largest context and best quality from Mistral's budget-friendly lineup.
TL;DR
- Price: $0.50/M input, $1.50/M output — 3:1 output ratio (lower than most)
- Context: 262K tokens — largest Mistral model, 2x Mistral Small 4
- Provider: Mistral (Paris, France) — European data sovereignty
- Best for: Long document analysis, agentic workflows, complex reasoning, enterprise chatbots
- vs Claude Sonnet 5: 4x cheaper input, 6.7x cheaper output, smaller context (262K vs 1M)
- vs Mistral Small 4: 3.3x more expensive but 2x context and stronger reasoning
Mistral Large 3 Pricing Breakdown
At $0.50 per million input tokens and $1.50 per million output tokens, Mistral Large 3 sits in the budget-to-mid tier — significantly cheaper than premium models (Claude Sonnet 5, GPT-5.4) while offering stronger reasoning and more context than ultra-budget options. Here's how costs scale:
| Monthly Volume | Input Cost | Output Cost | Total (50/50 split) |
|---|---|---|---|
| 1M tokens | $0.50 | $1.50 | $1.00 |
| 10M tokens | $5.00 | $15.00 | $10.00 |
| 100M tokens | $50.00 | $150.00 | $100.00 |
| 1B tokens | $500.00 | $1,500.00 | $1,000.00 |
At 100M tokens/month, Mistral Large 3 costs $100 — about 6x less than Claude Sonnet 5 ($600) and 2.7x less than GPT-5.4 ($275), while offering a 262K context window and strong reasoning capabilities.
Mistral Large 3 vs Other Budget Models
| Model | Input $/M | Output $/M | Context | Provider | Region |
|---|---|---|---|---|---|
| Qwen 3.7 Flash | $0.03 | $0.13 | 1M | Alibaba | China |
| GPT-5 nano | $0.05 | $0.40 | 128K | OpenAI | US |
| Gemini 2.5 Flash-Lite | $0.10 | $0.40 | 1M | US | |
| Mistral Small 4 | $0.15 | $0.60 | 128K | Mistral | EU |
| DeepSeek V4 Flash | $0.22 | $0.66 | 1M | DeepSeek | China |
| GPT-5.6 Luna | $0.20 | $1.20 | 1.05M | OpenAI | US |
| Gemini 2.5 Flash | $0.30 | $2.50 | 1M | US | |
| Codestral | $0.30 | $0.90 | 256K | Mistral | EU |
| Mistral Large 3 | $0.50 | $1.50 | 262K | Mistral | EU |
| Claude Haiku 4.5 | $1.00 | $5.00 | 200K | Anthropic | US |
Mistral Large 3 is 2x more expensive than Mistral Small 4 but offers 2x the context and stronger reasoning. It's 4x cheaper than Claude Haiku 4.5 on input and 3.3x cheaper on output — making it an excellent value for teams that need quality without premium pricing.
Mistral Model Family: Which One?
| Model | Input $/M | Output $/M | Context | Best For |
|---|---|---|---|---|
| Ministral 3 3B | $0.10 | $0.10 | 128K | Ultra-budget, edge, output-heavy |
| Ministral 3 8B | $0.15 | $0.15 | 128K | Symmetric pricing, simple tasks |
| Mistral Small 4 | $0.15 | $0.60 | 128K | General-purpose budget |
| Ministral 3 14B | $0.20 | $0.20 | 128K | Complex budget reasoning |
| Codestral | $0.30 | $0.90 | 256K | Code generation specialist |
| Mistral Large 3 | $0.50 | $1.50 | 262K | Best quality, long context |
| Magistral Small | $0.50 | $1.50 | 256K | Reasoning tasks |
| Mistral Medium 3.5 | $1.50 | $7.50 | 128K | Complex reasoning, enterprise |
Decision guide: Need the absolute cheapest? Ministral 3 3B. Need coding? Codestral. Need general-purpose quality on a budget? Mistral Small 4. Need long context and strong reasoning? Mistral Large 3. Need chain-of-thought reasoning? Magistral Small (same price). Need the absolute best Mistral quality? Mistral Medium 3.5.
Best Use Cases for Mistral Large 3
Long Document Analysis
262K context handles 100+ page contracts, research papers, and legal documents in a single pass — no chunking needed.
Agentic Workflows
Strong reasoning at budget pricing makes it ideal for multi-step agent pipelines that need to maintain context across tool calls.
Enterprise Chatbots
Extended conversation history without losing context. 262K tokens supports 50+ message threads with full history retention.
Research Synthesis
Analyze multiple papers, summarize findings, and generate insights across large document sets at 6x less than Claude Sonnet 5.
EU Compliance Workloads
GDPR-compliant processing for European enterprises that need both quality and data sovereignty — the best value from a European provider.
Data Pipeline Processing
Process and transform large datasets with strong reasoning. The 3:1 output ratio keeps costs predictable for output-heavy tasks.
Real-World Cost: 25K Document Analyses/Month
Suppose you're processing 25,000 long documents per month — each averaging 10K input tokens and generating 2K output tokens (summaries). That's 250M input tokens and 50M output tokens.
| Model | Input Cost | Output Cost | Total/Month |
|---|---|---|---|
| Qwen 3.7 Flash | $7.50 | $6.50 | $14.00 |
| Mistral Small 4 | $37.50 | $30.00 | $67.50 |
| Mistral Large 3 | $125.00 | $75.00 | $200.00 |
| Claude Haiku 4.5 | $250.00 | $250.00 | $500.00 |
| Claude Sonnet 5 | $500.00 | $500.00 | $1,000.00 |
Mistral Large 3 processes 25K long documents for $200/month — 5x cheaper than Claude Haiku 4.5 and 5x cheaper than Claude Sonnet 5, while offering comparable reasoning quality and a larger context window than Haiku.
Compare Mistral Large 3 Pricing
See how Mistral Large 3 stacks up against other models for your specific use case.
Frequently Asked Questions
Related Pages
- Mistral Small 4 Pricing — Budget option from Mistral at $0.15/$0.60
- Codestral Pricing — Mistral's code-specialized model at $0.30/$0.90
- Ministral 3 14B Pricing — Mistral's largest Ministral at $0.20/$0.20
- Claude Haiku 4.5 Pricing — Anthropic's budget model at $1.00/$5.00
- All Mistral Models — Full provider page with all Mistral pricing
- Cheapest AI APIs compared