Z.AI
availableShows if the model has enough results for an index.GLM-5.3-Flash
GLM-5.3-Flash is a reasoning model from Z.AI in the GLM-5 family. 41 benchmarks count toward its score, in 8 categories.
IndexOverall score. 50 is the middle.65.6 ±2.5
CoverageShare of the index weight with results.100%
SpeedOutput tokens per second.69/s
Input / 1MUS dollars per 1M input tokens.$0.15
Output / 1MUS dollars per 1M output tokens.$0.5
ContextMaximum tokens in one request.1.31M
EloLMArena rating and rank.1472 (#24)
50 is the middle of the board. The range shows the doubt in the index.
10,038 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
41 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | Math | 93.9% | 64.0 | max effort | — | Epoch AI |
| SWE-bench | Coding | 92.0% | 71.3 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis GPQA Diamond | Knowledge | 91.2% | 62.0 | — | — | Artificial Analysis |
| GPQA diamond | Knowledge | 90.2% | 61.5 | max effort | — | Epoch AI |
| CharXiv Reasoning | Multimodal | 89.4% | 67.9 | — | — | CharXiv authors |
| GPQA Diamond | Knowledge | 86.4% | 58.0 | — | 1 Sept 2026 | Vals AI |
| MMLU Pro | Knowledge | 86.1% | 56.1 | — | 1 Sept 2026 | Vals AI |
| MMMU Pro | Multimodal | 86.0% | 67.3 | — | 1 Sept 2026 | Vals AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 84.3% | 73.8 | — | — | DeepSeek-AI |
| Terminal-Bench 2.1 (provider run) | Agentic | 84.3% | 73.8 | — | — | DeepSeek-AI |
| LiveBench Mathematics | Math | 81.2% | 56.4 | — | 25 Jun 2026 | LiveBench |
| LiveCodeBench | Coding | 80.5% | 57.5 | — | 1 Sept 2026 | Vals AI |
| Multimodal Multi-disciplinary Video Understanding | Multimodal | 80.5% | — | — | — | MMVU benchmark maintainers |
| LiveBench Coding | Coding | 79.0% | 68.7 | — | 25 Jun 2026 | LiveBench |
| Toolathlon-Verified | Agentic | 78.4% | 75.9 | — | — | Moonshot AI |
| Chartography with image and code tools | Multimodal | 78.0% | — | — | — | Surge AI and Anthropic |
| LiveBench Reasoning | Reasoning | 77.6% | 63.8 | — | 25 Jun 2026 | LiveBench |
| LiveBench Language | Knowledge | 77.3% | 63.8 | — | 25 Jun 2026 | LiveBench |
| LiveBench Data Analysis | Reasoning | 76.4% | 62.1 | — | 25 Jun 2026 | LiveBench |
| EuroEval Portuguese | Multilingual | 74.1% | 94.8 | — | — | EuroEval |
| EuroEval French | Multilingual | 73.5% | 94.0 | — | — | EuroEval |
| EuroEval Italian | Multilingual | 73.3% | 93.7 | — | — | EuroEval |
| EuroEval Swedish | Multilingual | 72.8% | 93.2 | — | — | EuroEval |
| EuroEval Dutch | Multilingual | 69.9% | 89.6 | — | — | EuroEval |
| EuroEval Polish | Multilingual | 68.1% | 87.3 | — | — | EuroEval |
| EuroEval Spanish | Multilingual | 67.4% | 86.5 | — | — | EuroEval |
| EuroEval German | Multilingual | 63.6% | 81.7 | — | — | EuroEval |
| DeepSWE | Agentic | 63.4% | 69.9 | — | — | Datacurve AI |
| Terminal-Bench 2.1 | Agentic | 62.9% | 61.2 | — | 21 Sept 2026 | Vals AI |
| OfficeQA Pro | Multimodal | 62.4% | 74.7 | — | — | OfficeQA Pro authors |
| Artificial Analysis AutomationBench | Agentic | 60.4% | 72.5 | — | — | Artificial Analysis |
| OpenHarmony Bench v1.0 | Coding | 57.3% | 66.4 | — | — | OpenHarmony Bench authors |
| LiveBench Agentic Coding | Agentic | 56.8% | 70.1 | — | 25 Jun 2026 | LiveBench |
| NL2Repo | Coding | 56.3% | 69.3 | — | — | MiniMax |
| FrontierMath-Tiers-1-3-v2-Private | Math | 55.8% | 62.8 | max effort | — | Epoch AI |
| Humanity's Last Exam with tools | Agentic | 55.3% | 68.1 | — | — | DeepSeek-AI |
| BabyVision | Multimodal | 53.4% | — | — | — | Meta AI |
| LiveBench Instruction Following | Instruction | 52.8% | 43.7 | — | 25 Jun 2026 | LiveBench |
| IOI | Coding | 52.5% | 68.5 | — | 21 Sept 2026 | Vals AI |
| Medical Long Context Reasoning (MLCR-AA) | Reasoning | 51.1% | 82.5 | — | — | Wisedocs and Artificial Analysis |
| AutomationBench | Agentic | 48.8% | 95.0 | — | — | Moonshot AI |
| Artificial Analysis Tau3-Banking | Agentic | 47.2% | 76.7 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 41.8% | 74.7 | — | — | Artificial Analysis |
| SkillsBench | Coding | 40.2% | 57.0 | OpenHands | 11 Sept 2026 | Vals AI |
| Artificial Analysis EnterpriseOps-Gym | Agentic | 33.2% | 52.9 | — | — | Artificial Analysis |
| Vibe Code Bench v1.1 | Coding | 30.8% | 55.0 | OpenHands | 21 Sept 2026 | Vals AI |
| Agents' Last Exam | Agentic | 26.3% | 66.0 | — | — | DeepSeek-AI |
| ProofBench v1.1 | Math | 21.0% | 57.9 | — | 21 Sept 2026 | Vals AI |
| Code Migration | Coding | 20.5% | 58.0 | — | 21 Sept 2026 | Vals AI |
| Terminal-Bench 4.0 | Agentic | 19.7% | 73.3 | — | 21 Sept 2026 | Vals AI |
| FrontierMath-Tier-4-v2-Private | Math | 17.1% | 56.1 | max effort | — | Epoch AI |
| Vibe Code Bench 1-100 | Coding | 16.0% | 69.4 | OpenHands | 16 Sept 2026 | Vals AI |
| Chess Puzzles | Reasoning | 14.0% | 41.6 | max effort | — | Epoch AI |
| Agent Arena task outcome | Agentic | 8.2 | 76.7 | — | 15 Sept 2026 | LMArena |
| Mystery Game Puzzles | Reasoning | 8.0% | 42.4 | max effort | — | Epoch AI |
| Agent Arena steerability | Agentic | 0.1 | 67.5 | — | 15 Sept 2026 | LMArena |
| ProgramBench | Coding | 0.0% | — | — | 21 Sept 2026 | Vals AI |
| Agent Arena command recovery | Agentic | -4.6 | 62.1 | — | 15 Sept 2026 | LMArena |
41 benchmarks count, from 54 of 58 results. A grey row does not count. Too few models took that benchmark.