BlogLLM Benchmark Scores as Published, Side by Side
MMLU-Redux: Results as Published
Every result the catalog holds for MMLU-Redux: 39 models from 7 developers, read from 10 pages. Each number is as printed by the company named beside it; nothing is ranked or rescored. Models are listed by developer, newest first.
Area General Knowledge and ReasoningMaintainer University of Edinburgh (Aryo Pradipta Gema and co-authors)Introduced 2024Data version 2026103001, as of 2026-10-30
MMLU-Redux is a re-annotated subset of MMLU: 5,700 questions across all 57 subjects, checked by hand because about six percent of the original questions were found to be wrong. The score is accuracy.
MMLU-Redux
| Model | Value | Printed by | Setup as printed |
|---|---|---|---|
| GPT-5.2OpenAI, released 2025-12-11 | 95% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| GPT-5OpenAI, released 2025-08-07 | 95.3% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Effort: high; Tools: none |
| GPT-5 miniOpenAI, released 2025-08-07 | 93.7% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| gpt-oss-120bOpenAI, released 2025-08-05 | 91% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| GPT-4.1OpenAI, released 2025-04-14 | 92.4% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | - |
| o1-miniOpenAI, released 2024-09-12 | 86.7% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| GPT-4oOpenAI, released 2024-05-13 | 88% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| Claude Opus 4.5Anthropic, released 2025-11-24 | 95.6% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15; printed on 1 more page | - |
| Claude Sonnet 4.5Anthropic, released 2025-09-29 | 95.6% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Effort: thinking; Tools: none |
| Claude Opus 4Anthropic, released 2025-05-22 | 94.2% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking |
| Claude Sonnet 4Anthropic, released 2025-05-22 | 93.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | Effort: no extended thinking |
| Claude 3.5 SonnetAnthropic, released 2024-10-22 | 88.9% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| Gemma 4 26B A4BGoogle, released 2026-04-02 | 92.7% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15 | - |
| Gemma 4 31BGoogle, released 2026-04-02 | 93.7% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15; printed on 1 more page | - |
| Gemini 3 ProGoogle, released 2025-11-18 | 95.9% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Gemini 2.5 FlashGoogle, released 2025-05-20 | 90.6% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | - |
| DeepSeek-V4-Flash-BaseDeepSeek, released 2026-04-24 | 89.4% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15; printed on 1 more page | - |
| DeepSeek-V4-Pro-BaseDeepSeek, released 2026-04-24 | 90.8% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15; printed on 1 more page | - |
| DeepSeek-V3.2-BaseDeepSeek, released 2025-12-01 | 87.5% | DeepSeek, DeepSeek-V4-Pro (preview) model cardchecked 2026-10-15 | - |
| DeepSeek-V3.2-ExpDeepSeek, released 2025-09-29 | 93.7% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Tools: none |
| DeepSeek-R1-0528DeepSeek, released 2025-05-28 | 93.4% | DeepSeek, DeepSeek-R1-0528 model cardchecked 2026-10-15 | - |
| DeepSeek-V3-0324DeepSeek, released 2025-03-24 | 90.5% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | - |
| DeepSeek-R1DeepSeek, released 2025-01-20 | 92.9% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15; printed on 1 more page | - |
| DeepSeek-V3DeepSeek, released 2024-12-26 | 89.1% | DeepSeek, DeepSeek-R1 model cardchecked 2026-10-15 | - |
| Qwen3.6-27BAlibaba, released 2026-04-22 | 93.5% | Alibaba, Qwen3.6-27B model cardchecked 2026-10-15 | - |
| Qwen3.6-35B-A3BAlibaba, released 2026-04-16 | 93.3% | Alibaba, Qwen3.6-35B-A3B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3.5-122B-A10BAlibaba, released 2026-02-24 | 94% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15 | - |
| Qwen3.5-27BAlibaba, released 2026-02-24 | 93.2% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 2 more pages | - |
| Qwen3.5-35B-A3BAlibaba, released 2026-02-24 | 93.3% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3.5-397B-A17BAlibaba, released 2026-02-16 | 94.9% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3-235B-A22BAlibaba, released 2025-04-29 | 93.8% | Alibaba, Qwen3.5-122B-A10B model cardchecked 2026-10-15; printed on 1 more page | - |
| Qwen3-Max-ThinkingAlibaba | 92.8% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Kimi K2.5Moonshot AI, released 2026-01-27 | 94.5% | Alibaba, Qwen3.5-397B-A17B model cardchecked 2026-10-15 | - |
| Kimi K2 ThinkingMoonshot AI, released 2025-11-06 | 94.4% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Effort: thinking; Tools: none |
| Kimi K2 Instruct 0905Moonshot AI, released 2025-09-05 | 92.7% | Moonshot AI, Kimi K2 Thinking release pagechecked 2026-10-15 | Tools: none |
| Kimi K2 BaseMoonshot AI, released 2025-07-11 | 90.2% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| Kimi K2 InstructMoonshot AI, released 2025-07-11 | 92.7% | Moonshot AI, Kimi K2 release pagechecked 2026-10-15 | - |
| MiMo-V2.5 BaseXiaomi, released 2026-04-27 | 89.8% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
| MiMo-V2.5-Pro BaseXiaomi, released 2026-04-27 | 92.8% | Xiaomi, Xiaomi MiMo-V2.5-Pro model cardchecked 2026-10-15 | Harness: 5-shot |
Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a 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.