Mistral Medium 3.5
MistralOpen weightsModified MITReleased Apr 29, 2026
On the arena.ai Agents leaderboard, Mistral Medium 3.5 has a Net Improvement of -12.65% (—, rank #47, 95% interval -14.33%–-10.97%). Mistral Medium 3.5 ranks #122 on the LMArena Text leaderboard (style control) with an Arena score of 1426.4 (±6.4, 11,791 votes). API price: $1.50 input / $7.50 output per 1M tokens. Context: 262K. Nearest rivals: MiMo V2.5 (1433.9) above and Claude Haiku 5.5 (1424.8) below.
Quick answer
- Agents leaderboard
- #47 of 37 · -12.65% success
- Text (LMArena)
- #122 · 1426.4
- 17NAS overall
- #36 of 36
- API price (in / out)
- $1.50 / $7.50 per 1M tokens
- Closest rival
- MiMo V2.5 · 1433.9
- Cheaper alternative
- MiniMax M3: Text 1439.2, output $1.20 (84% cheaper)
- Data updated
- Oct 29, 2026 · synced daily
Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter
Key metrics
Rank history
LMArena rank, last 14 days#117 on Oct 16, #122 now; best #117 in this period.
Models ranked nearby
Sorted by Arena score| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| 56 |  GPT-5.2 | 1437.3 | – | $14.00 | – |
| 57 |  DeepSeek V4 Flash | 1435.7 | – | $0.28 | – |
| 58 |  MiMo V2.5 | 1433.9 | – | $0.27 | – |
| 59 | Mistral Medium 3.5This page | 1426.4 | – | $7.50 | 157 |
| 60 |  Claude Haiku 5.5 | 1424.8 | 36.2 | $0.50 | 236 |
| 61 | DeepSeek V3.2 | 1424.6 | – | $0.42 | – |
| 62 |  GLM-4.6 | 1424.3 | – | $2.20 | – |
More from Mistral
| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| – | Mistral Large | – | – | $12.00 | – |
Pricing
- Official API list price: $1.50 input and $7.50 output per 1M tokens (USD).
- At a typical 3:1 input-to-output ratio, the blended cost is about $3.00 per 1M tokens.
- Among 67 models with an output price, it is the #35 cheapest.
- The cheapest OpenRouter route charges $1.50 input / $7.50 output.
Sources & methodology
On the arena.ai Agents leaderboard, Mistral Medium 3.5 has a Net Improvement of -12.65% (—, rank #47, 95% interval -14.33%–-10.97%).
- arena.ai Agents — agent Net Improvement, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
- LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
- Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
- OpenRouter — Route prices, context length and listing date.
Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).
Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.
FAQ
How does Mistral Medium 3.5 do on the Agents leaderboard?
On the arena.ai Agents leaderboard, Mistral Medium 3.5 has a Net Improvement of -12.65% (—, rank #47, 95% interval -14.33%–-10.97%).
What rank is Mistral Medium 3.5 right now?
Mistral Medium 3.5 is #59 on our leaderboard of 63 scored models (Arena score 1426.4); its raw LMArena rank is #122.
What does Mistral Medium 3.5 cost via API?
Input $1.50 / output $7.50 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $3.00 per 1M tokens.
How fast is Mistral Medium 3.5?
Artificial Analysis measures about 157 output tokens per second, 0.9 s to the first token and 14 s to the first answer token; it is a reasoning model, so latency includes thinking time.
How close is Mistral Medium 3.5 to the models ranked next to it?
It is 7.5 points behind MiMo V2.5 (1433.9) and 1.6 points ahead of Claude Haiku 5.5 (1424.8); gaps smaller than the confidence interval (±6.4) are not significant.
Is Mistral Medium 3.5 open source? Can it be self-hosted?
Mistral Medium 3.5 is an open-weights model (license: Modified MIT); you can download the weights and self-host.
What is the context length of Mistral Medium 3.5?
262K (262,144 tokens, as listed on OpenRouter).
Mistral Medium 3.5 vs MiMo V2.5: which is better?
On the LMArena Text leaderboard Mistral Medium 3.5 scores 1426.4 and MiMo V2.5 scores 1433.9 (a 7.5-point gap). On price, Mistral Medium 3.5 costs $7.50 and MiMo V2.5 $0.27 per 1M output tokens.
Is there a cheaper model close to Mistral Medium 3.5?
Yes: MiniMax M3 scores 1439.2 on the Text leaderboard (Mistral Medium 3.5: 1426.4) and costs $1.20 per 1M output tokens, about 84% less than Mistral Medium 3.5 ($7.50).
Where does the Mistral Medium 3.5 data come from and how often is it updated?
Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Oct 29, 2026.