Mistral
Mistral Medium 3.5, Mistral Small 4, Voxtral TTS, Mistral OCR 4 — European sovereign AI
Mistral AI is a Paris-based lab offering both proprietary and open-weight models. The September 2026 lineup includes Mistral Medium 3.5 (frontier, 128K context, vision), Mistral Small 4 (efficient), Voxtral TTS (text-to-speech), and Mistral OCR 4 (document intelligence). Mistral Studio is their platform for building, testing, and deploying AI agents. Vibe is their AI agent product for long-horizon work. Check mistral.ai/technology/#pricing for current rates.
The only frontier lab offering European data sovereignty with open-weight models — plus Mistral OCR 4 for document intelligence, Voxtral TTS for voice, and Mistral Studio for agentic AI.
- Mistral Medium 3.5 with vision — competitive frontier pricing
- Mistral OCR 4 for document intelligence and PDF processing
- Voxtral TTS for voice applications
- European data sovereignty — GDPR-native, data stays in EU
- Several open-weight models available for self-hosting
- Mistral Studio for building, testing, and deploying AI agents
- Intelligence scores trail OpenAI and Anthropic at frontier tier
- Smaller ecosystem than OpenAI
- Vibe and Studio products are newer and less battle-tested
Mistral Models
| Model | Modality | Tier | Price | Cached /1M | TTFT | Throughput | Context | Intel. | Caps | |
|---|---|---|---|---|---|---|---|---|---|---|
mistral-small-latest Mistral Small | Chat | budget | $0.10/ $0.30 in / out per 1M EST Jul 2026 | — | 280ms | 250 t/s | 128K | 50 | ||
mistral-small-3.2 Mistral Small | Chat | budget | $0.10/ $0.30 in / out per 1M EST Jul 2026 | — | 250ms | 280 t/s | 131K | 55 | ||
devstral-small-2505 Devstral | Chat | budget | $0.10/ $0.30 in / out per 1M EST Jul 2026 | — | 280ms | 260 t/s | 131K | 58 | ||
mistral-embed Mistral Embed | Embedding | budget | $0.10/ $0.00 in / out per 1M EST Jul 2026 | — | 80ms | 0 t/s | 8K | 0 | ||
mistral-small-4 Mistral Small | Chat | budget | $0.15/ $0.60 in / out per 1M EST Oct 2026 | — | 240ms | 270 t/s | 131K | 58 | ||
mistral-nemo Mistral Nemo | Chat | budget | $0.15/ $0.15 in / out per 1M EST Jul 2026 | — | 200ms | 350 t/s | 131K | 48 | ||
pixtral-12b Pixtral | Chat | efficient | $0.15/ $0.15 in / out per 1M EST Jul 2026 | — | 300ms | 180 t/s | 128K | 68 | ||
codestral-latest Codestral | Chat | efficient | $0.30/ $0.90 in / out per 1M EST Jul 2026 | — | 350ms | 200 t/s | 262K | 62 | ||
devstral-medium Devstral | Chat | efficient | $0.40/ $2.00 in / out per 1M EST Jul 2026 | — | 400ms | 150 t/s | 128K | 74 | ||
mistral-large-latest Mistral Large | Chat | advanced | $0.50/ $1.50 in / out per 1M EST Jul 2026 | — | 550ms | 130 t/s | 131K | 70 | ||
magistral-small-2506 Magistral | Chat | efficient | $0.50/ $1.50 in / out per 1M EST Jul 2026 | — | 800ms | 150 t/s | 131K | 65 | ||
mistral-large-4 Mistral Large | Chat | advanced | $0.68/ $2.09 in / out per 1M EST Oct 2026 | — | 500ms | 140 t/s | 131K | 72 | ||
mistral-medium-3 Mistral Medium | Chat | efficient | $1.50/ $7.50 in / out per 1M EST Oct 2026 | — | 400ms | 180 t/s | 131K | 65 | ||
mistral-medium-3.5 Mistral Medium | Chat | efficient | $1.50/ $7.50 in / out per 1M EST Oct 2026 | — | 380ms | 185 t/s | 131K | 68 | ||
magistral-medium-2506 Magistral | Chat | advanced | $2.00/ $5.00 in / out per 1M EST Jul 2026 | — | 1.5s | 80 t/s | 131K | 75 | ||
pixtral-large-2411 Pixtral | Chat | advanced | $2.00/ $6.00 in / out per 1M EST Jul 2026 | — | 600ms | 110 t/s | 131K | 72 | ||
mistral-large Mistral Large | Chat | advanced | $2.00/ $6.00 in / out per 1M EST Jul 2026 | — | 600ms | 100 t/s | 128K | 76 |
Mistral — Frequently Asked Questions (FAQs)
Similar Providers
Mistral Reviews
All LLM Providers
Full comparison tableMistral LLM pricing overview
Mistral is a proprietary AI lab that develops and hosts its own large language models. Mistral AI is a Paris-based lab offering both proprietary and open-weight models. The September 2026 lineup includes Mistral Medium 3.5 (frontier, 128K context, vision), Mistral Small 4 (efficient), Voxtral TTS (text-to-speech), and Mistral OCR 4 (document intelligence). Mistral Studio is their platform for building, testing, and deploying AI agents. Vibe is their AI agent product for long-horizon work. Check mistral.ai/technology/#pricing for current rates. Common use cases include Coding, European compliance, Open-source, Cost-efficiency. Headquartered in Paris, France, founded 2023. All models are accessible via a REST API compatible with standard OpenAI-style request formats, enabling drop-in integration with most LLM frameworks and orchestration tools.
Mistral vs other LLM providers
Mistral competes with OpenAI, Anthropic, Google, Groq, Together AI, Mistral, Cohere, and other inference API providers across dimensions of price, throughput, latency, context window, and model intelligence. The only frontier lab offering European data sovereignty with open-weight models — plus Mistral OCR 4 for document intelligence, Voxtral TTS for voice, and Mistral Studio for agentic AI. Use the LLM provider comparison tool to see Mistral token pricing, latency, and throughput side-by-side with any other provider. The full LLM pricing table shows all providers ranked by input token cost, output token cost, and throughput in a single sortable view.
Why choose Mistral?
Mistral stands out because: The only frontier lab offering European data sovereignty with open-weight models — plus Mistral OCR 4 for document intelligence, Voxtral TTS for voice, and Mistral Studio for agentic AI.. The API currently offers 17 models, spanning 5 advanced, 6 efficient, 6 budget tiers. For teams running high-volume inference workloads, prompt caching and batch API endpoints can reduce effective input token costs by 50–90% — check the context window cost guide for a full breakdown of caching economics.
Understanding Mistral token pricing
Mistral charges separately for input (prompt) and output (completion) tokens, priced per 1M tokens in USD. Output tokens are typically 3–5× more expensive than input tokens due to the compute cost of autoregressive generation. The cheapest {meta.name} input rate is currently $0.10/1M tokens. Prices shown are sourced from Mistral's public pricing page and updated every 15 minutes. Need help estimating your spend? Read the LLM API cost calculator guide — it covers tokens, context windows, prompt caching, and batch discounts with worked examples for RAG, chat history, and document processing workloads.
Mistral context window and model capabilities
Context window size directly affects both capability and cost — every token in the context window is charged as an input token. Mistral's largest context window is 262K tokens. For RAG and document processing workloads, longer context windows enable richer retrieval but increase per-call costs proportionally. Prompt caching — where supported — stores the KV state of repeated prefixes and charges 75–90% less for cache hits, making it the most impactful cost optimization for applications with consistent system prompts or retrieved documents. See the LLM context window cost guide for a full analysis of how context length affects your API bill.
Self-hosted vs managed inference: when Mistral makes sense
Managed inference APIs like Mistral eliminate infrastructure overhead — no GPU provisioning, driver management, or model serving stack to maintain. The trade-off is cost at scale: a single H100 at ~$2.50/hr can serve ~500K tokens/min of Llama 3.3 70B, which at Mistral API rates would cost significantly more per token. The break-even point depends on your request volume, latency requirements, and engineering capacity. For teams processing fewer than ~10M tokens/day, managed APIs are almost always cheaper when total cost of ownership is considered. Above that threshold, self-hosted inference on rented GPU compute typically wins on unit economics. Read the cheapest GPU cloud guide for a full break-even analysis. Historical Mistral token price data is available in the LLM price history charts.
Compare Mistral With Another LLM Provider
Compare Mistral vs other providers
Side-by-side token pricing, latency, throughput, and context window. View all LLM provider comparisons →
Looking for alternatives?
Compare Mistral with other LLM API providers side by side.