BlogLLM Benchmark Scores as Published, Side by Side
MiMo-V2.6-Pro: Benchmark Scores as Published
Every result the catalog holds for MiMo-V2.6-Pro: 20 tests in 6 areas, from 5 pages. 17 printed by Xiaomi itself, the rest by the company named beside each number. Each number is as printed; nothing is ranked or rescored.
Developer XiaomiReleased 2026-09-21Context window 1MInput Text, imageOutput TextWeights Open (MIT)Data version 2026103001, as of 2026-10-30
Knowledge Work
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| AutomationBench | 53.1% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| JobBench | 62 | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| GDPval-AA v2.1 | 1673 | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| AA-Briefcase | 1516 | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
Agentic Coding
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Terminal-Bench 2.1 | 89.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| DeepSWE v1.1 | 71.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| Terminal-Bench 4.0 | 34.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| ProgramBench Almost@1 | 26.5% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo Code Bench (Xiaomi internal) | 63.2% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| MiMo VisualCoding (Xiaomi internal) | 72.3% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
Science and Math
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| SciCode-Verified | 91.9% | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | Attempt budget: pass@1 (n=6) |
Computer Use
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| OSWorld-Verified | 82% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
Cybersecurity
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| CyberGym | 94% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| ExploitBench | 47.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| ExploitGym | 17.8% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19; printed on 1 more page | - |
| SEC-Bench Pro | 66.3% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19; printed on 1 more page | - |
| CyberGym-E2E | 79% | Mistral AI, Mistral Large 4 (preview)checked 2026-10-15 | - |
| MiMo Cyber Bench (Xiaomi internal) | 80.2% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
Tool Use and Workflows
| Test | Value | Printed by | Setup as printed |
|---|---|---|---|
| Agents' Last Exam (pass rate) | 31.6% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
| Toolathlon Verified | 76.9% | Xiaomi, Xiaomi MiMo-V2.6-Pro-RL model cardchecked 2026-10-19 | - |
Price
API list prices in US dollars per one million tokens, as printed on the developer's pricing page on the day checked. Cost per task is measured by Vals AI.
| Price | Value | Printed by | Setup as printed |
|---|---|---|---|
| Input | $0.435 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Cached input | $0.0036 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Output | $0.87 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Input, batch | $0.2175 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Cached input, batch | $0.0018 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Output, batch | $0.435 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Input, ultrafast | $4.35 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Cached input, ultrafast | $0.036 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Output, ultrafast | $8.7 | Xiaomi, Xiaomi MiMo API pricingchecked 2026-10-19 | Region: international |
| Cost per taskVals AI, USD, lower is better | $1.04 | Vals AI, Vibe Code Bench v1.1 leaderboardchecked 2026-10-27 | Harness: OpenHands |
More from Xiaomi
Every number here is read from the same catalog file as the benchmark table, data version 2026103001. Where one page printed the model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.