MiMo V2.5 Pro
XiaomiOpen weightsMITReleased Apr 22, 2026
On the arena.ai Agents leaderboard, MiMo V2.5 Pro has a task success rate of -7.67% (—, rank #46, 95% interval -8.69%–-6.64%). MiMo V2.5 Pro ranks #49 on the LMArena Text leaderboard (style control) with an Arena score of 1467.3 (±3.6, 75,721 votes). Artificial Analysis Intelligence Index: 26.0. API price: $0.44 input / $0.87 output per 1M tokens. Context: 1.05M. Nearest rivals: Claude Opus 4.5 (1469.6) above and GPT-5.6 Terra (1466.4) below.
Quick answer
- Agents leaderboard
- #46 of 48 · -7.67% success
- Text (LMArena)
- #49 · 1467.3
- 17NAS overall
- #46 of 84
- API price (in / out)
- $0.44 / $0.87 per 1M tokens
- Closest rival
- Claude Opus 4.5 · 1469.6
- Cheaper alternative
- GLM-5.3 Flash: Text 1473.8, output $0.50 (43% cheaper)
- Data updated
- Oct 16, 2026 · synced daily
Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter
Key metrics
Reasoning effort comparison
Artificial Analysis measurements per reasoning effort of the same model| Effort | Intelligence | Coding | tok/s | TTFT (s) | First answer (s) |
|---|---|---|---|---|---|
| reasoningUsed in table | 26.0 | 60.2 | – | – | – |
| non-reasoning | 18.3 | – | – | – | – |
Models ranked nearby
Sorted by Arena score| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| 36 |  Claude Sonnet 4.6 | 1472.1 | 24.7 | $15.00 | – |
| 37 |  Grok 4.20 | 1471.5 | 25.7 | $2.50 | – |
| 38 | Claude Opus 4.5 | 1469.6 | 23.7 | $25.00 | – |
| 39 | MiMo V2.5 ProThis page | 1467.3 | 26.0 | $0.87 | – |
| 40 |  GPT-5.6 Terra | 1466.4 | 42.1 | $12.00 | 116 |
| 41 | Grok 4.5 | 1465.7 | 38.8 | $6.00 | – |
| 42 |  GLM-5.1 | 1465.2 | 26.1 | $4.07 | – |
More from Xiaomi
| # | Model | Arena | Intel. | Output $ | tok/s |
|---|---|---|---|---|---|
| 24 | MiMo-V2.6-Pro | 1479.2 | 46.3 | $0.87 | 50 |
| 54 | MiMo-V2.6-Flash | 1450.9 | 37.9 | $0.28 | 51 |
| 64 | MiMo V2.5 | 1433.8 | 21.8 | $0.28 | – |
Pricing
- Official API list price: $0.44 input and $0.87 output per 1M tokens (USD).
- At a typical 3:1 input-to-output ratio, the blended cost is about $0.54 per 1M tokens.
- Among 79 models with an output price, it is the #10 cheapest.
- The cheapest OpenRouter route charges $0.44 input / $0.87 output.
Sources & methodology
On the arena.ai Agents leaderboard, MiMo V2.5 Pro has a task success rate of -7.67% (—, rank #46, 95% interval -8.69%–-6.64%).
- arena.ai Agents — agent task success rate, 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 MiMo V2.5 Pro do on the Agents leaderboard?
On the arena.ai Agents leaderboard, MiMo V2.5 Pro has a task success rate of -7.67% (—, rank #46, 95% interval -8.69%–-6.64%).
What rank is MiMo V2.5 Pro right now?
MiMo V2.5 Pro is #39 on our leaderboard of 69 scored models (Arena score 1467.3); its raw LMArena rank is #49.
What does MiMo V2.5 Pro cost via API?
Input $0.44 / output $0.87 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 $0.54 per 1M tokens.
How close is MiMo V2.5 Pro to the models ranked next to it?
It is 2.3 points behind Claude Opus 4.5 (1469.6) and 0.9 points ahead of GPT-5.6 Terra (1466.4); gaps smaller than the confidence interval (±3.6) are not significant.
Is MiMo V2.5 Pro open source? Can it be self-hosted?
MiMo V2.5 Pro is an open-weights model (license: MIT); you can download the weights and self-host.
What is the context length of MiMo V2.5 Pro?
1.05M (1,050,000 tokens, as listed on OpenRouter).
MiMo V2.5 Pro vs Claude Opus 4.5: which is better?
On the LMArena Text leaderboard MiMo V2.5 Pro scores 1467.3 and Claude Opus 4.5 scores 1469.6 (a 2.3-point gap, within the confidence interval, so not significant). On price, MiMo V2.5 Pro costs $0.87 and Claude Opus 4.5 $25.00 per 1M output tokens.
Is there a cheaper model close to MiMo V2.5 Pro?
Yes: GLM-5.3 Flash scores 1473.8 on the Text leaderboard (MiMo V2.5 Pro: 1467.3) and costs $0.50 per 1M output tokens, about 43% less than MiMo V2.5 Pro ($0.87).
Where does the MiMo V2.5 Pro 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 16, 2026.