CC-Bench-V2 Multi-Step Chained Tasks

Only in system cards

Labs’ own numbers All independent runs

Agents complete chains of tasks mined from merged pull requests, where each step changes the codebase for the next. Tests check every step.

In-house Z.ai evaluation. Pass@1 on individual tasks.

Made by: Zhipu AI (Z.ai)
Released: February 2026
Kind: Fixing real repositories
Level: Expert
Scored: Checked automatically
Answers: Agent task
Tasks: Kept private

Strongest model tested: Claude Opus 4.5 #61 on the capability index
Last in a system card: February 2026
Last model added: February 2026
Scores refreshed: November 12th, 2026

Paper arxiv.org

Model scored in lab cards

System card reports it, from 1 lab

GLM-5 leads with 52.3%

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

Claude Opus 4.5 leads with 61.6% across all independent runs

Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.

Bars Table

52.3 GLM-5

1 GLM-5 52.3%

Source: Labs’ model cards

1 Claude Opus 4.5 61.6%
2 GLM-5 52.3%

Source: Official CC-Bench-V2 Multi-Step Chained Tasks leaderboard

#	Model	Score	Setting	Date	Source
1	GLM-5	52.3%		Feb 2026	Lab model card

#	Model	Score	Setting	Date	Source
1	Claude Opus 4.5	61.6%		Nov 2025	Official leaderboard
2	GLM-5	52.3%		Feb 2026	Official leaderboard

The benchmark's own leaderboard has 2 models. Switch the scores above to rank them together.

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Reported by the lab in its model card

Sources: Official CC-Bench-V2 Multi-Step Chained Tasks leaderboard and labs’ model cards

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
Feb 17, 2026	GLM-5 technical report	52.3	CC-Bench-V2 Long-horizon Chained Tasks, Pass@1 (Table 8)	Capabilities

One lab claim next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. 1 of them match, which usually means one side copied the other. Every benchmark.

Model	Lab says	Board says	Gap
GLM-5	52.3% CC-Bench-V2 Long-horizon Chained Tasks, Pass@1 (Table 8)	52.3% Leaderboard	Same