Mistral Medium 3
by Mistral · mistralai/mistral-medium-3 · #162 cheapest of 325 paid models
Prices updated Jul 22, 2026, 2:49 AM UTC · refreshed hourly
Mistral Medium 3 is a high-performance enterprise-grade language model designed to deliver frontier-level capabilities at significantly reduced operational cost. It balances state-of-the-art reasoning and multimodal performance with 8× lower cost...
Compare vs:Claude Sonnet 5Claude Opus 4.6Claude Haiku 4.5GPT-5.6 Sol
Pricing
Input
$0.40
per 1M tokens
Output
$2.00
per 1M tokens
Blended (3:1)
$0.80
3 input : 1 output
Cache read
$0.04
per 1M cached tokens
Cache write
—
per 1M tokens
Cache & batch economics
Effective prices for Mistral Medium 3 with prompt caching.
| Cache hit rate | Effective blended $/1M | Example request* | vs no cache |
|---|---|---|---|
| 0% | $0.80 | $0.0428 | — |
| 50% | $0.665 | $0.0248 | −42% |
| 90% | $0.557 | $0.0104 | −76% |
*Example: 100K sticky context + 2K new input + 1K output. Batch discount is an approximate 50% off list rates (available for OpenAI / Anthropic / Google — not flagged for this creator). Confirm on the vendor’s pricing page.
What a request costs
| Workload | Input tokens | Output tokens | Cost / request | Cost / 1K requests |
|---|---|---|---|---|
| Short chat message | 500 | 300 | $0.0008 | $0.8 |
| Document summary | 8,000 | 1,000 | $0.0052 | $5.20 |
| Codebase question (RAG) | 30,000 | 2,000 | $0.016 | $16.00 |
| Long-context analysis | 150,000 | 5,000 | $0.07 | $70.00 |
Performance
Intelligence Index
12.5
AA composite quality
Coding Index
—
Math Index
30.3
Agentic Index
—
Output speed
50 tok/s
median
Time to first token
0.55s
median
Time to first answer
0.55s
after reasoning tokens
Artificial Analysis benchmarks
| Benchmark | Score |
|---|---|
| MMLU-Pro | 76.0% |
| GPQA Diamond | 57.8% |
| Humanity's Last Exam | 4.3% |
| LiveCodeBench | 40.0% |
| SciCode | 33.1% |
| MATH-500 | 90.7% |
| AIME | 44.0% |
| AIME 2025 | 30.3% |
| IFBench | 39.3% |
| AA-LCR (Long Context Reasoning) | 28.0% |
| Terminal-Bench Hard | 3.8% |
| τ²-Bench (Telecom) | 24.3% |
Design Arena Elo
| Category | Elo |
|---|---|
| Website | 1,107 |
| UI components | 1,069 |
| Data viz | 1,069 |
| 3D | 1,141 |
| Game dev | 1,077 |
| Code categories | 1,103 |
Specs
Context window
131K
tokens
Max output
—
tokens
Input modalities
text, image, file
Tokenizer
Mistral