Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work
Specification
mistral listing
Context
1M
Max output
262K
Released
2026-10-06
Knowledge cutoff
—
Retires
—
Open weights
no
Input
text, image, pdf
Output
text
Price
US dollars per million tokens · mistral list price
Input
$0.70
Output
$2.10
Cache read
$0.07
Quality
9 benchmarks · 2 sources
Composite
53rd
percentile of 344 scored models
Rank
161 / 344
±0.08 sd
Evidence
9 × 2
3 or more benchmarks, from 2 or more sources
Effort range
—
one configuration only
reasoning70th3/4 bench
math74th3/7 bench
knowledge15th1/1 bench
general27th1/1 bench
Every score
one row per source and configuration — nothing averaged away
| Benchmark | Score | Configuration | Source | Run |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 98.3 ±0.6 | high | Epoch AI | 2026-10-12 |
| GPQA diamond | 88.1 ±1.9 | high | Epoch AI | 2026-10-12 |
| FrontierMath-Tiers-1-3-v2 | 58.6 ±2.9 | high | Epoch AI | 2026-10-12 |
| Chess Puzzles | 33.0 ±4.7 | high | Epoch AI | 2026-10-15 |
| LiveBench | 72.8 | high · 23 runs | LiveBench | 2026-06-25 |
| Mystery Game Puzzles | 17.0 ±3.8 | high | Epoch AI | 2026-10-13 |
| FrontierMath-Tier-4-v2 | 26.8 ±7.0 | high | Epoch AI | 2026-10-12 |
| SimpleQA Verified | 20.0 ±1.3 | high | Epoch AI | 2026-10-09 |
| Furniture Assembly | 20.8 ±5.2 | high | Epoch AI | 2026-10-14 |
The leading figure for a benchmark is the median across its configurations, so one heroic high-effort run cannot set the number.
Available from 10 hosts
| Host | In $/M | Out $/M | Cache rd | Cache wr | Context | Output | Retires |
|---|---|---|---|---|---|---|---|
| mistralvendor | $0.70 $1.36 | $2.10 $4.18 | $0.07 $0.14 | — | 1M | 262K 1M | — |
| nano-gpt | $0.68 | $2.09 | $0.07 | — | 524K | 262K | — |
| ollama-cloud | $0.68 | $2.09 | $0.07 | — | 1M | 262K | — |
| opencode | $0.68 | $2.09 | $0.07 | — | 524K | 262K | — |
| requesty | $0.68 | $2.09 | $0.07 | — | 524K | 262K | — |
| vercel | $0.70 | $2.10 | $0.07 | — | 1M | 262K | — |
| commandcode | $1.36 | $4.18 | $0.14 | — | 524K | 262K | — |
| llmgateway | $1.36 | $4.18 | $0.14 | — | 1M | 262K | — |
| llmgateway-providers | $1.36 | $4.18 | $0.14 | — | 1M | 262K | — |
| lowrouter · auto | $1.4367 | $4.4156 | $0.1479 | — | 524K | 262K | — |
Price history
append-only observations · a listing writes a row only when its price moves
| Date | Host | Field | $/M |
|---|---|---|---|
| 2026-10-07 | mistral | Cache read | $0.07 |
| 2026-10-23 | mistral | Cache read | $0.14 |
| 2026-10-29 | mistral | Cache read | $0.14 |
| 2026-10-30 | mistral | Cache read | $0.07 |
| 2026-10-07 | vercel | Cache read | $0.07 |
| 2026-10-21 | vercel | Cache read | $0.14 |
| 2026-10-30 | vercel | Cache read | $0.07 |
| 2026-10-07 | mistral | Input | $0.68 |
| 2026-10-23 | mistral | Input | $1.36 |
| 2026-10-29 | mistral | Input | $1.36 |
| 2026-10-30 | mistral | Input | $0.70 |
| 2026-10-07 | vercel | Input | $0.68 |
| 2026-10-21 | vercel | Input | $1.36 |
| 2026-10-30 | vercel | Input | $0.70 |
| 2026-10-07 | mistral | Output | $2.09 |
| 2026-10-23 | mistral | Output | $4.18 |
| 2026-10-29 | mistral | Output | $4.18 |
| 2026-10-30 | mistral | Output | $2.10 |
| 2026-10-07 | vercel | Output | $2.09 |
| 2026-10-21 | vercel | Output | $4.18 |
| 2026-10-30 | vercel | Output | $2.10 |
| 2026-10-08 | ollama-cloud | Cache read | $0.14 |
| 2026-10-22 | ollama-cloud | Cache read | $0.07 |
| 2026-10-08 | ollama-cloud | Input | $1.36 |
| 2026-10-22 | ollama-cloud | Input | $0.68 |
| 2026-10-08 | ollama-cloud | Output | $4.18 |
| 2026-10-22 | ollama-cloud | Output | $2.09 |
| 2026-10-10 | lowrouter · auto | Cache read | $0.0732 |
| 2026-10-22 | lowrouter · auto | Cache read | $0.1479 |
| 2026-10-10 | lowrouter · auto | Input | $0.7112 |
| 2026-10-22 | lowrouter · auto | Input | $1.4367 |
| 2026-10-10 | lowrouter · auto | Output | $2.1859 |
| 2026-10-22 | lowrouter · auto | Output | $4.4156 |
Other listings of this model
same model, different key
More from Mistral
most-hosted first