Xiaomiopen-weight
Released: 2026-09-21Artificial AnalysisMiMo-V2.6-Flash
OpenRouter ranked #8 foundation model processing 4.04T tokens weekly. Deployed by xiaomi.
Intelligence Index
26.7/ 100
Artificial Analysis Intelligence Index (independent composite)Throughput Speed
46tok/sec
Output streaming throughputTime to First Answer
39s
Time to first answer token, including any thinkingCost to Run the Index
0.13USD
Artificial Analysis cost of the full Index runMeasured results · Artificial Analysis
Independent evaluations of MiMo-V2.6-Flash. Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.
- Intelligence Index
- 26.7
- Cost to run the Index
- $0.13
- Time per task
- 47.8 min
- Omniscience index
- -10.8
- Reasoning model
- Yes
- Cached-input price
- $0.0028 / 1M
Reasoning & knowledge
- AA-Omniscience Accuracy27.2%
- CritPt12.0%
- CritPt v2 (Hard)29.5%
- HLE-Diamond13.7%
- Humanity's Last Exam35.1%
Coding & terminal agents
- FrontierCode 1.142.1%
- FrontierCode-AA 2.021.6%
- SciCode51.3%
- Terminal-Bench 4.022.7%
- Terminal-Bench-Science 0.14.3%
Agentic work
- AA-Briefcase v1.1 (Elo)1489 Elo
- AutomationBench-AA64.0%
- GDPval-AA (normalized Elo)1615 Elo
Long context & documents
Multimodal & science
- Code2Video-AA (Elo)1052 Elo
- MMMU-Pro73.1%
Health & safety
Headline Evaluation Breakdown
Six independently measured evaluations (Artificial Analysis)Reasoning · CritPt v2 (Hard)29.5%
Coding & agents · Terminal-Bench 4.022.7%
Long context · AA-LCR74.3%
Knowledge · Humanity's Last Exam35.1%
Sourced Benchmark Evaluations (16)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| FrontierCode | Active | 42.1% | 98 | 2026-11-15 | independent | Source ↗ |
| Humanity's Last Exam | Active | 35.1% | 93 | 2026-11-15 | independent | Source ↗ |
| MMMU-Pro | Active | 73.1% | 93 | 2026-11-15 | independent | Source ↗ |
| SciCode | Active | 51.3% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench 4.0 | Active | 22.7% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench-Science 0.1 | Active | 4.3% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-LCR (Long Context Reasoning) | Active | 74.3% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt | Active | 12% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt v2 (Hard) | Active | 29.5% | 93 | 2026-11-15 | independent | Source ↗ |
| HLE-Diamond | Active | 13.7% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Accuracy) | Active | 27.2% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Non-Hallucination) | Active | 47.8% | 93 | 2026-11-15 | independent | Source ↗ |
| FrontierCode-AA 2.0 | Active | 21.6% | 93 | 2026-11-15 | independent | Source ↗ |
| GDP.pdf | Active | 9% | 93 | 2026-11-15 | independent | Source ↗ |
| MLCR-AA (Medical Long Context Reasoning) | Active | 11.7% | 93 | 2026-11-15 | independent | Source ↗ |
| AutomationBench-AA | Active | 64.1% | 93 | 2026-11-15 | independent | Source ↗ |
Local Hardware Execution (Ollama)Hardware Compatible
# 1. Pull and execute model locally with Ollama
ollama run mimo-v2-6-flash
# 2. Or invoke via local OpenAI-compatible endpoint
curl http://localhost:11434/v1/chat/completions \
-H "Content-Type: application/json" \
-d '{
"model": "mimo-v2-6-flash",
"messages": [{"role": "user", "content": "Analyze reasoning chains on AIME 2026"}]
}'Evaluation & Sourcing Notes
Scores listed for MiMo-V2.6-Flash represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.