Compare
Four models, one page
Pick the models you are actually choosing between. Every row is scaled within itself, so a bar means something next to its neighbours and nothing across rows.
| Specification | MiMo-V2.5-ProXiaomi | Claude Opus 5Anthropic |
|---|---|---|
| Released | Apr 22, 2026 | Jul 24, 2026 |
| Context window | 1.05M | 1M |
| Max output | 131K | 128K |
| Input / 1M | $0.435 | $5 |
| Output / 1M | $0.87 | $25 |
| Cost per task | $0.09 | $2.25 |
| Arena Elo | 1,314 | 1,393 |
| Serving providers | 7 | 5 |
| Parameters | 1023.2B | Undisclosed |
| Licence | mit | Proprietary |
| Capabilities |
|
|
Metrics side by side
Each row is scaled to the largest value in that row — bars compare within a row, never across rows
- MiMo-V2.5-Pro
- Claude Opus 5
Intelligence Index
MiMo-V2.5-Pro42.9Claude Opus 563.1Coding Index
MiMo-V2.5-Pro60.2Claude Opus 578.0Agentic Index
MiMo-V2.5-Pro29.5Claude Opus 559.2Output speed
MiMo-V2.5-Pro73 t/sClaude Opus 587 t/sContext window
MiMo-V2.5-Pro1.05MClaude Opus 51MLatency · lower is better
MiMo-V2.5-Pro479msClaude Opus 51.12sBlended price / 1M · lower is better
MiMo-V2.5-Pro$0.544Claude Opus 5$10Cost per task · lower is better
MiMo-V2.5-Pro$0.09Claude Opus 5$2.25
Rows marked “lower is better” still draw a longer bar for a larger number — read the value, not just the length. Arena Elo is in the specification table above instead: it has no meaningful zero, so a bar would flatten the gaps.
View as table
| Metric | MiMo-V2.5-Pro | Claude Opus 5 |
|---|---|---|
| Intelligence Index | 42.9 | 63.1 |
| Coding Index | 60.2 | 78.0 |
| Agentic Index | 29.5 | 59.2 |
| Output speed | 73 t/s | 87 t/s |
| Context window | 1.05M | 1M |
| Latency · lower is better | 479ms | 1.12s |
| Blended price / 1M · lower is better | $0.544 | $10 |
| Cost per task · lower is better | $0.09 | $2.25 |
Evaluation scores
Percentage correct on a common 0–100% scale
- MiMo-V2.5-Pro
- Claude Opus 5
GPQA Diamond
MiMo-V2.5-Pro86.6%Claude Opus 593.2%Humanity's Last Exam
MiMo-V2.5-Pro35.7%Claude Opus 554.9%SciCode
MiMo-V2.5-Pro50.2%Claude Opus 555.7%τ²-bench
MiMo-V2.5-Pro94.2%Claude Opus 542.1%Terminal-Bench Hard
MiMo-V2.5-Pro43.2%Claude Opus 5not measuredAA-LCR long context
MiMo-V2.5-Pro77.7%Claude Opus 575.7%
A missing bar means that evaluation was not run for that model — it is not a zero.
View as table
| Evaluation | MiMo-V2.5-Pro | Claude Opus 5 |
|---|---|---|
| GPQA Diamond | 86.6% | 93.2% |
| Humanity's Last Exam | 35.7% | 54.9% |
| SciCode | 50.2% | 55.7% |
| τ²-bench | 94.2% | 42.1% |
| Terminal-Bench Hard | 43.2% | — |
| LiveCodeBench | — | — |
| AA-LCR long context | 77.7% | 75.7% |
| AIME 2025 | — | — |
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).