Agentic & Tool Use benchmark # METR Time Horizons leaderboard > METR Time Horizons results for 32 AI models, led by Claude Mythos Preview at 85.2%. What the benchmark measures, who runs it, and a source for every score. - Canonical page: https://noometry.com/benchmarks/metr-time-horizons - Last updated: 2026-10-10 - Title: METR Time Horizons Leaderboard (October 2026): Scores by Model As of October 2026, Claude Mythos Preview has the highest published METR Time Horizons score on Noometry at 85.2%, out of 32 models with results. Last verified October 10, 2026 ## About METR Time Horizons METR's measure of how long a task (in human expert time) a model can complete with 50% reliability. - **Category:** [Agentic & Tool Use](https://noometry.com/best/agentic) - **Introduced:** 2025 - **Format:** Software tasks - **Unit:** Percent (random guessing ≈ 0%) - **Official site:** [metr.org](https://metr.org) ## Top 15 models Top models on METR Time Horizons 1. Claude Mythos Preview 85.2% 2. Claude Opus 4.6 78.9% 3. Gemini 3.1 Pro Preview 77% 4. GPT-5.2 75.3% 5. Claude Opus 4.5 75% 6. GPT-5.3 Codex 74.5% 7. GPT-5.4 74.3% 8. Gemini 3 Pro 71% 9. GPT-5.1-Codex 70.8% 10. GPT-5 69.6% 11. Claude Sonnet 4.5 67.4% 12. Claude Opus 4.1 66.8% 13. Grok 4 66.6% 14. o3 65.4% 15. Claude Opus 4 63.9% 16. 5060708090 Sponsored placements are available on pages like this one. [Advertise on Noometry](https://noometry.com/advertise) ## All results METR Time Horizons results by model | # | Model | Provider | Score | Setting | Source | Date | | --- | --- | --- | --- | --- | --- | --- | | 1 | [Claude Mythos Preview](https://noometry.com/models/claude-mythos-preview) | [Anthropic](https://noometry.com/providers/anthropic) | 85.2% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 2 | [Claude Opus 4.6](https://noometry.com/models/claude-opus-4-6) | [Anthropic](https://noometry.com/providers/anthropic) | 78.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 3 | [Gemini 3.1 Pro Preview](https://noometry.com/models/gemini-3-1-pro-preview) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 77% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 4 | [GPT-5.2](https://noometry.com/models/gpt-5-2) | [OpenAI](https://noometry.com/providers/openai) | 75.3% | high | [Epoch AI](https://epoch.ai/benchmarks) | | | 5 | [Claude Opus 4.5](https://noometry.com/models/claude-opus-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 75% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 6 | [GPT-5.3 Codex](https://noometry.com/models/gpt-5-3-codex) | [OpenAI](https://noometry.com/providers/openai) | 74.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 7 | [GPT-5.4](https://noometry.com/models/gpt-5-4) | [OpenAI](https://noometry.com/providers/openai) | 74.3% | xhigh | [Epoch AI](https://epoch.ai/benchmarks) | | | 8 | [Gemini 3 Pro](https://noometry.com/models/gemini-3-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 71% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 9 | [GPT-5.1-Codex](https://noometry.com/models/gpt-5-1-codex) | [OpenAI](https://noometry.com/providers/openai) | 70.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 10 | [GPT-5](https://noometry.com/models/gpt-5) | [OpenAI](https://noometry.com/providers/openai) | 69.6% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 11 | [Claude Sonnet 4.5](https://noometry.com/models/claude-sonnet-4-5) | [Anthropic](https://noometry.com/providers/anthropic) | 67.4% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 12 | [Claude Opus 4.1](https://noometry.com/models/claude-opus-4-1) | [Anthropic](https://noometry.com/providers/anthropic) | 66.8% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 13 | [Grok 4](https://noometry.com/models/grok-4) | [xAI](https://noometry.com/providers/xai) | 66.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 14 | [o3](https://noometry.com/models/o3) | [OpenAI](https://noometry.com/providers/openai) | 65.4% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 15 | [Claude Opus 4](https://noometry.com/models/claude-opus-4) | [Anthropic](https://noometry.com/providers/anthropic) | 63.9% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 16 | [o4-mini](https://noometry.com/models/o4-mini) | [OpenAI](https://noometry.com/providers/openai) | 63.9% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 17 | [Claude Sonnet 4](https://noometry.com/models/claude-sonnet-4) | [Anthropic](https://noometry.com/providers/anthropic) | 62% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 18 | [Claude 3.7 Sonnet](https://noometry.com/models/claude-3-7-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 60% | 16K | [Epoch AI](https://epoch.ai/benchmarks) | | | 19 | [Kimi K2 (Jul 2025)](https://noometry.com/models/kimi-k2) | [Moonshot AI](https://noometry.com/providers/moonshot) | 59.2% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 20 | [gpt-oss-120b](https://noometry.com/models/gpt-oss-120b) | [OpenAI](https://noometry.com/providers/openai) | 56.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 21 | [Gemini 2.5 Pro](https://noometry.com/models/gemini-2-5-pro) | [![](/logos/google.svg) Google](https://noometry.com/providers/google) | 55.4% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 22 | [DeepSeek-R1](https://noometry.com/models/deepseek-r1) | [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 53.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 23 | [o1](https://noometry.com/models/o1) | [OpenAI](https://noometry.com/providers/openai) | 51.1% | medium | [Epoch AI](https://epoch.ai/benchmarks) | | | 24 | [DeepSeek-V3](https://noometry.com/models/deepseek-v3) | [![](/logos/deepseek.svg) DeepSeek](https://noometry.com/providers/deepseek) | 49.6% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 25 | [Claude 3.5 Sonnet](https://noometry.com/models/claude-3-5-sonnet) | [Anthropic](https://noometry.com/providers/anthropic) | 45.2% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 26 | [GPT-4o](https://noometry.com/models/gpt-4o) | [OpenAI](https://noometry.com/providers/openai) | 40.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 27 | [GPT-4 Turbo](https://noometry.com/models/gpt-4-turbo) | [OpenAI](https://noometry.com/providers/openai) | 36.7% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 28 | [GPT-4](https://noometry.com/models/gpt-4) | [OpenAI](https://noometry.com/providers/openai) | 36.1% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 29 | [Qwen2.5 72B Instruct](https://noometry.com/models/qwen2-5-72b-instruct) | [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 35.8% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 30 | [Qwen2-72B](https://noometry.com/models/qwen2-72b) | [![](/logos/alibaba.svg) Alibaba (Qwen)](https://noometry.com/providers/alibaba) | 29.9% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 31 | [Claude 3 Opus](https://noometry.com/models/claude-3-opus) | [Anthropic](https://noometry.com/providers/anthropic) | 29.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | | 32 | [GPT-3.5-turbo](https://noometry.com/models/gpt-3-5-turbo) | [OpenAI](https://noometry.com/providers/openai) | 21.5% | | [Epoch AI](https://epoch.ai/benchmarks) | | ## Compare the leaders - [Claude Mythos Preview vs Claude Opus 4.6](https://noometry.com/compare/claude-mythos-preview-vs-claude-opus-4-6) - [Claude Mythos Preview vs Gemini 3.1 Pro Preview](https://noometry.com/compare/claude-mythos-preview-vs-gemini-3-1-pro-preview) - [Claude Mythos Preview vs GPT-5.2](https://noometry.com/compare/claude-mythos-preview-vs-gpt-5-2) - [Claude Mythos Preview vs Claude Opus 4.5](https://noometry.com/compare/claude-mythos-preview-vs-claude-opus-4-5) - [Claude Opus 4.6 vs Gemini 3.1 Pro Preview](https://noometry.com/compare/claude-opus-4-6-vs-gemini-3-1-pro-preview) - [Claude Opus 4.6 vs GPT-5.2](https://noometry.com/compare/claude-opus-4-6-vs-gpt-5-2) ## Other agentic & tool use benchmarks - [Terminal-Bench](https://noometry.com/benchmarks/terminal-bench) - [APEX-Agents](https://noometry.com/benchmarks/apex-agents) - [Berkeley Function Calling Leaderboard](https://noometry.com/benchmarks/bfcl) - [OSWorld 2.0](https://noometry.com/benchmarks/osworld-2) - [GDPval](https://noometry.com/benchmarks/gdpval) - [Remote Labor Index](https://noometry.com/benchmarks/remote-labor-index) - [TheAgentCompany](https://noometry.com/benchmarks/the-agent-company) - [τ²-bench Airline](https://noometry.com/benchmarks/tau2-airline) - [τ²-bench Banking](https://noometry.com/benchmarks/tau2-banking) - [τ²-bench Retail](https://noometry.com/benchmarks/tau2-retail) - [τ²-bench Telecom](https://noometry.com/benchmarks/tau2-telecom) - [Cybench](https://noometry.com/benchmarks/cybench) ## Frequently asked questions ### What does METR Time Horizons measure? METR's measure of how long a task (in human expert time) a model can complete with 50% reliability. ### Which model has the highest METR Time Horizons score? As of October 2026, Claude Mythos Preview has the highest published METR Time Horizons score on Noometry at 85.2%, out of 32 models with results. ### What is the best open-weight model on METR Time Horizons? Kimi K2 (Jul 2025) has the highest METR Time Horizons accuracy among open-weight models at 59.2%, ranking 19 of 32 overall. ### Cite this page Noometry. (2026). METR Time Horizons leaderboard. Retrieved October 10, 2026, from https://noometry.com/benchmarks/metr-time-horizons Quote Noometry with a link back to this page. It is also available in [Markdown](https://noometry.com/md/benchmarks/metr-time-horizons.md).