CC-Bench
Claude Code runs realistic frontend, backend, tooling and data coding tasks in fresh containers. Humans compare models head-to-head (win rate).
Lab-created in-house eval (Zhipu/Z.ai). V1.x questions and trajectories on HF, but task environments are not released. CC-Bench-V2 (GLM-5) is internal.
- Made by
- Zhipu AI (Z.ai)
- Released
- July 2025
- Kind
- Building whole projects
- Level
- Expert
- Scored
- Ranked by votes
- Answers
- Agent task
- Strongest model tested
- Claude Sonnet 4 #112 on the capability index
- Last in a system card
- August 2025
- Last model added
- September 2025
- Scores refreshed
- November 12th, 2026
Claude Sonnet 4 leads with 41.9% among models released by September 2025
Best score per model from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in September 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- Claude Fable 5.2
-
Gemini 4 Carbon - GPT-6 Astra
- Claude Opus 5.5
- and 16 more
Source: Official CC-Bench leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 4 | 41.9% | reference GLM-4.6 | May 2025 | Official leaderboard |
| 2 | GLM-4.5 | 36.5% | reference GLM-4.6 | Aug 2025 | Official leaderboard |
| 3 |  DeepSeek-V3.1-Terminus | 27% | reference GLM-4.6 | Sep 2025 | Official leaderboard |
| 4 | Kimi K2 Sep 2025 | 14.9% | reference GLM-4.6 | Sep 2025 | Official leaderboard |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official CC-Bench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official CC-Bench leaderboard
Labs report it only in system cards
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Aug 8, 2025 | GLM-4.5 technical report | CC-Bench head-to-head, win rate vs Claude Sonnet 4 40.4%, vs Kimi K2 53.9%, vs Qwen3-Coder 80.8% (Figure 12, text), no overall score. Tool calling success 90.6% (Figure 13) | Capabilities |
More in Coding
- SWE-bench Verified In use 500 human-validated GitHub issues from 12 Python repos. Agent writes a patch that must pass hidden tests. Claude Opus 4.7 83.5%
- LiveCodeBench v6 In use Competitive programming problems collected after model cutoffs from LeetCode, AtCoder, Codeforces. Scored pass@1 on hidden tests. DeepSeek-V4-Pro 93.5%
- SWE-bench Multilingual In use 300 GitHub-issue tasks across nine programming languages in SWE-bench format. Patch graded by repository tests. Claude Fable 5.2 96.6%
- DeepSWE In use 113 original long-horizon engineering tasks across 91 repos and five languages, written from scratch. Hand-written verifiers accept alternative solutions. GPT-6 Astra 74.1%
- SciCode In use Scientist-curated research coding problems split into subproblems across natural sciences. Code checked against test cases.
Seed2.1 Pro 59.8% - Codeforces rating In use Model solves recent Codeforces contest problems. An Elo-style rating is computed from simulated contest performance. DeepSeek V4.1 Flash 3,471