Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Loading page…
Preferred languages
More languages supported by Google Translate
Open the selector to load the current language catalog.
Translations are provided by Google Translate.
Z.AI · Open Weight · rank 116 · bench-align-v5
GLM-5 is a general-purpose language model. It accepts text and produces text.
GLM-5 by Z.AI scores 45.8 overall, costs $1.00 per 1M input tokens and $3.20 per 1M output tokens, and carries a context window of 200K tokens — observed 2026-11-21.
Seven axes from the ranking source. A missing axis is drawn as a gap.
Each axis shows the ranking source’s score and, beside it, BenchLM’s separate verified score for that axis. The two are published side by side and are not interchangeable.
Every published price component, including cache reads and writes.
10 conversations per day × 8 messages × 22 active days, 1200 input and 400 output tokens per message, no cache. This uses the same calculator as the cost simulator, so an unavailable applicable rate makes the example unavailable too.
Lifecycle events the source associates with this model, plus what this profile does not claim.
That is not evidence the model has no lifecycle plan — only that this source published none.
Measured values with the date they were observed. Nothing is inferred from a sibling model or a provider claim.
Time to first token: latency to the first answer chunk (Artificial Analysis, via BenchLM). Reasoning models include thinking time, so values can run to tens or hundreds of seconds.
Evidence key: Observed
| Measurement | Value | Observed | Last good | Evidence |
|---|---|---|---|---|
| Time to first token | 45.24 s | 2026-11-21 | 2026-11-21 | |
| Throughput | 71 tok/s | 2026-11-21 | 2026-11-21 |
Regional or per-endpoint measurements appear only when the API supplies them; none are modelled here.
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| GPQA Graduate-Level Google-Proof Q&A | 86 | #53 | 448 questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| GPQA-D GPQA Diamond | 86.0 | #61 | Graduate-level science questions | Graduate level | GPQA: A Graduate-Level Google-Proof Q&A Benchmark |
| SuperGPQA SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines | 66.8 | #13 | 285 disciplines | Graduate level | SuperGPQA: Scaling LLM Evaluation Across 285 Graduate Disciplines |
| MMLU-Pro Massive Multitask Language Understanding Professional | 85.7 | #17 | Multiple subjects | Professional level | MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark |
| HLE Humanity's Last Exam | 30.5 | #52 | Expert-level questions | Frontier expert level | Humanity's Last Exam |
| Artificial Analysis Intelligence Index Artificial Analysis Intelligence Index | 27.9 | #70 | Cross-benchmark intelligence index | Display-only external reference | Artificial Analysis |
| AA-GPQA Diamond Artificial Analysis GPQA Diamond | 82.0 | #96 | Graduate-level science questions | Graduate-level science reasoning | Artificial Analysis GPQA Diamond Benchmark Leaderboard |
| AA-HLE Artificial Analysis Humanity's Last Exam | 29.3 | #87 | Expert-level questions | Frontier expert reasoning | Artificial Analysis Humanity's Last Exam Benchmark Leaderboard |
| AA-Omniscience Index Artificial Analysis Omniscience Index | 0.3 | #78 | Knowledge questions | Broad factual knowledge | AA-Omniscience: Knowledge and Hallucination Benchmark |
| AA-Omniscience Accuracy Artificial Analysis Omniscience Accuracy | 26.3 | #108 | Knowledge questions | Broad knowledge | Artificial Analysis model benchmarks |
| AA-Omniscience Hallucination Rate Artificial Analysis Omniscience Hallucination Rate | 35.3 | #174 | Knowledge questions | Factuality | Artificial Analysis model benchmarks |
| MMLU-Pro (Arcee) MMLU-Pro first-party comparison snapshot | 85.8 | #3 | Professional academic QA | Professional level | Trinity-Large-Thinking: Scaling an Open Source Frontier Agent |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 Terminal-Bench 2.0 | 56.2 | #47 | Terminal-based software tasks | Professional software engineering | Terminal-Bench 2.0 |
| SWE-bench Verified Software Engineering Benchmark Verified | 77.8 | #33 | 500 verified issues | Professional software engineering | SWE-bench: Can Language Models Resolve Real-World GitHub Issues? |
| SWE-Rebench SWE-Rebench | 62.8 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| AIME25 (Arcee) AIME25 first-party comparison snapshot | 93.3 | #4 | 15 problems | High school olympiad level | Trinity-Large-Thinking: Scaling an Open Source Frontier Agent |
| AIME26 AIME 2026 | 95.8 | #9 | Competition math problems | Olympiad-style mathematics | Qwen3.6 launch benchmarks |
| HMMT Feb 2025 Harvard-MIT Mathematics Tournament February 2025 | 97.5 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| LongBench v2 LongBench v2 | 60.8 | #14 | Long-context tasks | Hard long-context | LongBench v2 |
| AI-Needle AI-Needle | 63.3 | #4 | Long-context retrieval | Long-context memory | Qwen3.6 launch benchmarks |
| AA-LCR Artificial Analysis Long Context Reasoning | 75.7 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| IFEval Instruction-Following Eval | 92.6 | #16 | 541 prompts across 25 instruction types | Instruction precision | Instruction-Following Evaluation for Large Language Models |
| AA-IFBench Artificial Analysis IFBench | 72.3 | #39 | Verifiable instruction constraints | Instruction precision | Artificial Analysis IFBench Benchmark Leaderboard |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| MMLU-ProX MMLU-ProX | 83.1 | #6 | Multilingual professional QA | Professional multilingual | MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluation |
| NOVA-63 NOVA-63 | 55.1 | #7 | Broad multilingual evaluation | Broad multilingual capability | Qwen3.6 launch benchmarks |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Terminal-Bench 2.0 Terminal-Bench 2.0 | 56.2 | #49 | Terminal-based software tasks | Professional software engineering | Terminal-Bench 2.0 |
| BrowseComp BrowseComp | 62 | #53 | Research questions requiring browsing | Hard web research | BrowseComp |
| HLE w/ tools Humanity's Last Exam with tools | 50.4 |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| Design Arena Website Design Arena Website Elo | 1236 | #54 | Website generation comparisons | Design and website generation | OpenRouter Grok 4.3 benchmarks |
| Benchmark | Value | Rank | Tasks | Difficulty | Provenance |
|---|---|---|---|---|---|
| BinaryAudit BinaryAudit (backdoor detection in compiled binaries) | 19.3 | #12 | 19 backdoor and timebomb detection tasks on stripped binaries, three runs each (57 runs); 6 tooling and 8 clean-binary tasks reported alongside | Reverse engineering without source code or symbols | We hid backdoors in ~40 MB binaries and asked AI + Ghidra to find them (Quesma) |
| #2 |
| Fresh GitHub issues (rolling window) |
| Professional software engineering |
| SWE-Rebench: Contamination-Free Evaluation of Software Engineering Agents |
| SWE-bench Pro SWE-bench Pro | 55.1 | #57 | 1,865 repository problems | Long-horizon professional engineering | SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks? |
| SWE Multilingual SWE Multilingual | 73.3 | #33 | Multilingual software-engineering tasks | Professional software engineering | MiniMax M2.7: Early Echoes of Self-Evolution |
| NL2Repo NL2Repo | 35.9 | #31 | Natural language to repository tasks | System-level software comprehension | MiniMax M2.7: Early Echoes of Self-Evolution |
| React Native Evals React Native Evals | 74.8 | #11 | React Native app implementation tasks | Production mobile app engineering | React Native Evals |
| SWE-bench Verified* SWE-bench Verified (mini-swe-agent-v2) | 72.8 | #3 | Repository task completion | Professional software engineering | Trinity-Large-Thinking: Scaling an Open Source Frontier Agent |
| #1 |
| Competition math problems |
| Olympiad-style mathematics |
| Qwen3.6 launch benchmarks |
| HMMT Nov 2025 Harvard-MIT Mathematics Tournament November 2025 | 96.9 | #1 | Competition math problems | Olympiad-style mathematics | Qwen3.6 launch benchmarks |
| HMMT Feb 2026 Harvard-MIT Mathematics Tournament February 2026 | 86.4 | #16 | Competition math problems | Olympiad-style mathematics | Qwen3.6 launch benchmarks |
| MMAnswerBench MMAnswerBench | 82.5 | #6 | Multimodal math questions | Advanced mathematical reasoning | Qwen3.6 launch benchmarks |
| FrontierMath v2 (Tiers 1-3) FrontierMath v2 Tiers 1-3 | 16.434 | #78 | 285 private advanced mathematics problems | From olympiad-plus to early research mathematics | FrontierMath v2 benchmark hub |
| FrontierMath v2 (Tier 4) FrontierMath v2 Tier 4 | 2.100 | #67 | 41 private extreme-difficulty mathematics problems | Research-level mathematics requiring hours or days of expert work | FrontierMath Tier 4 v2 leaderboard |
| Long-context reasoning tasks |
| Long-context reasoning |
| Artificial Analysis model benchmarks |
| CritPt Critical Physics Tasks | 2.0 | #109 | Research-level physics questions | Research-level physics reasoning | CritPt Benchmark Leaderboard |
| Expert questions with tool use |
| Frontier tool-augmented reasoning |
| DeepSeek-V4 Technical Report |
| APEX-Agents-AA APEX-Agents-AA | 14.5 | #26 | 452 professional-services agent tasks | Long-horizon workplace agent tasks | APEX-Agents-AA Benchmark Leaderboard |
| Gert Labs Gert Labs Composite Game Benchmark | 50.99 | #20 | Novel game environments | Agentic coding and decision-making | Gert Labs rankings |
| CyberGym CyberGym | 43.2 | #31 | 1,507 vulnerability analysis instances | Real-world cybersecurity | CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale |
| MCP Atlas MCP Atlas | 31.1 | #54 | Tool-integrated agent tasks | Advanced tool use | Introducing GPT-5.4 mini and nano |
| Toolathlon Toolathlon | 38 | #19 | Multi-tool workflows | Advanced tool use | Introducing GPT-5.4 mini and nano |
| τ²-bench results τ²-Bench Tool-Agent-User Evaluation | 98.2 | #7 | Airline, retail, and telecom customer-service task sets | Dual-control customer-service workflows | τ²-Bench: Evaluating Conversational Agents in a Dual-Control Environment |
| Claw-Eval Claw-Eval | 57.7 | #32 | 300 tasks, 2,159 rubrics | Real-world general, multi-turn, and native multimodal agent execution | Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agents |
| QwenClawBench QwenClawBench | 54.1 | #6 | Real-world agent workflows | Broad real-world agentic execution | Qwen3.6 launch benchmarks |
| τ³-bench results τ³-Bench Tool-Agent-User Evaluation | 65.6 | #14 | Corrected customer-service tasks plus knowledge and voice evaluation modes | Long-horizon, multimodal, and knowledge-aware tool use | Official τ³-bench repository and release notes |
| DeepPlanning DeepPlanning | 14.6 | #6 | Travel planning and constrained shopping | Constrained agent planning | DeepPlanning: Benchmarking Long-Horizon Agentic Planning with Verifiable Constraints |
| MCP-Tasks MCP-Tasks | 60.8 | #4 | MCP-integrated tool tasks | Advanced MCP workflows | Qwen3.6 launch benchmarks |
| WideResearch WideResearch | 69.8 | #13 | Open-ended research tasks | Broad research-agent workflows | Qwen3.6 launch benchmarks |