CL-bench leaderboard
As of 6 Nov 2026, the top score on CL-bench is 27.9% by GPT-5.4 (OpenAI), ahead of GPT-5.1 at 23.7%, across 21 model configurations with a published result (Epoch AI Benchmarking Hub, result dated 5 Mar 2026). Tasks where the model must apply new rules or facts supplied in the prompt rather than prior knowledge. CL-bench is shown for reference only and does not count toward the BenchLeader Index.
Key facts (as of 6 Nov 2026)
- GPT-5.4 holds the top CL-bench score at 27.9% (Epoch AI Benchmarking Hub, 5 Mar 2026).
- GPT-5.1 is second on CL-bench at 23.7% (Epoch AI Benchmarking Hub, 13 Nov 2025).
- CL-bench: Continual-learning style tasks that require using information given in context (Epoch AI Benchmarking Hub).
- 21 model configurations have a published CL-bench result on BenchLeader (BenchLeader, 6 Nov 2026).
- CL-bench does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).
Continual-learning style tasks that require using information given in context.
- Published by
- Epoch AI Benchmarking Hub
- Category
- Reasoning
- Index weight
- Reference only
- Models
- 21
- Data as of
- 6 Nov 2026
CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.
What the test looks like
Tasks where the model must apply new rules or facts supplied in the prompt rather than prior knowledge.
How it is scored
Overall score, as published by CL-bench.
What to keep in mind
New; small set.
- 1GPT-5.4 (xhigh)27.9%
- 2GPT-5.1 (high)23.7%
- 3Grok 4.2022.2%
- 4Claude Opus 4.521.1%
- 5GPT-5.121.1%
- 6Gemini 3.1 Pro20.8%
- 7Claude Opus 4.620.7%
- 8Qwen3.6 Plus20.3%
- 9Qwen3.5 Plus19.8%
- 10Kimi K2.519.3%
- 11GLM-518.7%
- 12GPT-5.218.2%
- 13GPT-5.2 (high)18.1%
- 14o3 (high)17.8%
- 15Kimi K2 Thinking (thinking)17.6%
| # | ||||
|---|---|---|---|---|
| 1 |  GPT-5.4xhigh | 27.9% | 63.2 | 2026-03-05 |
| 2 | GPT-5.1high | 23.7% | 57.7 | 2025-11-13 |
| 3 |  Grok 4.20 | 22.2% | – | 2026-02-17 |
| 4 |  Claude Opus 4.5 | 21.1% | 57.9 | 2025-11-24 |
| 5 | GPT-5.1 | 21.1% | 56.8 | 2025-11-13 |
| 6 |  Gemini 3.1 Pro | 20.8% | 60.5 | 2026-02-19 |
| 7 | Claude Opus 4.6 | 20.7% | 62.3 | 2026-02-05 |
| 8 |  Qwen3.6 Plus | 20.3% | 59.1 | 2026-03-31 |
| 9 | Qwen3.5 Plus | 19.8% | – | 2026-02-16 |
| 10 |  Kimi K2.5open ↗ | 19.3% | 54.2 | 2026-01-27 |
| 11 |  GLM-5open ↗ | 18.7% | 54.8 | 2026-02-11 |
| 12 | GPT-5.2 | 18.2% | 57.9 | 2025-12-11 |
| 13 | GPT-5.2high | 18.1% | 56.0 | 2025-12-11 |
| 14 | o3high | 17.8% | 53.8 | 2025-04-16 |
| 15 | Kimi K2 Thinkingthinkingopen ↗ | 17.6% | 53.2 | 2025-11-06 |
| 16 | GLM 4.7open ↗ | 15.9% | 51.7 | 2025-12-22 |
| 17 | Gemini 3 Pro | 15.8% | 60.2 | 2025-11-18 |
| 18 |  Mimo v2 Pro | 15.7% | 59.1 | – |
| 19 | Qwen3 Maxmax | 14.5% | 54.3 | 2025-09-24 |
| 20 |  DeepSeek V3.2thinkingopen ↗ | 13.2% | 56.6 | 2025-09-29 |
| 21 |  MiniMax M2.5open ↗ | 11.4% | 53.6 | 2026-02-12 |
More reasoning benchmarks · All LLM benchmarks
Related benchmarks
Top head-to-heads on CL-bench
Cite as: BenchLeader, “CL-bench leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/cl_bench.
CL-bench: questions
- What does CL-bench measure?
- Tasks where the model must apply new rules or facts supplied in the prompt rather than prior knowledge. Scores are reported in percent of tasks solved; higher is better.
- What is the top score on CL-bench?
- As of 6 Nov 2026, the highest score on CL-bench is 27.9% by GPT-5.4 (result dated 5 Mar 2026), ahead of GPT-5.1 at 23.7%. Scores come from Epoch AI Benchmarking Hub and are re-read daily.
- Which AI model leads CL-bench?
- GPT-5.4 leads CL-bench with 27.9% as of 6 Nov 2026 (result dated 5 Mar 2026), ahead of GPT-5.1 at 23.7%.
- How many models have CL-bench results?
- 21 model configurations have a CL-bench result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
- Who runs CL-bench and how often is it updated?
- CL-bench is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does CL-bench count toward the BenchLeader Index?
- No. CL-bench is shown for reference but left out of the composite index.