GLM-5.3 的 CyberGym · GLM-5.3模型卡;unknown 官方出处。GLM-5.3 · max · 84.5。见《GLM-5.3 official model card》Benchmark table;CyberGym;GLM-5.3目标列。证据 ev-gmc-glm-5-3-83。
GLM-5.3 · CyberGym(GLM-5.3模型卡;unknown),GLM-5.3 · max配置,pass-at-1 84.5(具体成绩公布时间未确认)。见《GLM-5.3 official model card》Benchmark table;CyberGym;GLM-5.3目标列。 查看成绩条件
| Benchmark | GLM-5.3 | GLM-5.2 | Kimi K3 | DeepSeek-V4 Pro-0813 | Qwen3.8-Max | Opus 4.8 | Fable 5 (w/ fallback) | GPT-5.6 Sol |
| CyberGym | **84.5** | 77.2 | 80.0 | 83.3 | 78.5 | 78.1 | 83.8 | 83.6 |
- **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating.
GLM-5.3 supports controlling the thinking budget through the `reasoning_effort` parameter, which accepts three levels: `low`, `high`, and `max`. It defaults to `max` if not passed (or if set to any other value). To use `low` or `high`, pass them explicitly. For benchmark and leaderboard reproduction, keep the default `max`.
The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating.
条件
GLM-5.3目标列;The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating. 安全、内部子项或跨测试聚合仅保留原始披露,不进入常规展示。
GLM-5.3 · CyberGym (GLM-5.3模型卡;unknown) 84.5 · Main text evidence (ev-gmc-glm-5-3-83)
Official source for CyberGym · GLM-5.3模型卡;unknown in GLM-5.3. GLM-5.3 · max · 84.5. See “GLM-5.3 official model card” at Benchmark table;CyberGym;GLM-5.3目标列. Evidence ev-gmc-glm-5-3-83.
GLM-5.3 · CyberGym (GLM-5.3模型卡;unknown), configuration GLM-5.3 · max, pass-at-1 84.5 (exact score publication date unconfirmed). See “GLM-5.3 official model card” at Benchmark table;CyberGym;GLM-5.3目标列. View score conditions
| Benchmark | GLM-5.3 | GLM-5.2 | Kimi K3 | DeepSeek-V4 Pro-0813 | Qwen3.8-Max | Opus 4.8 | Fable 5 (w/ fallback) | GPT-5.6 Sol |
| CyberGym | **84.5** | 77.2 | 80.0 | 83.3 | 78.5 | 78.1 | 83.8 | 83.6 |
- **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating.
GLM-5.3 supports controlling the thinking budget through the `reasoning_effort` parameter, which accepts three levels: `low`, `high`, and `max`. It defaults to `max` if not passed (or if set to any other value). To use `low` or `high`, pass them explicitly. For benchmark and leaderboard reproduction, keep the default `max`.
The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating.
Conditions
GLM-5.3目标列;The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **CyberGym**: We evaluate GLM-5.3 in Claude Code 2.1.207 (max reasoning effort, no web tools with `temperature=1.0`, `top_p=1.0`, `max_new_tokens=128000`). All evaluations are under unlimited timeout per task and results are single-run Pass@1 over 1,507 tasks. To simulate real-world usage scenarios, we place the agent inside the task container. We also remove all Git-related information and apply a domain whitelist (allowing only essential domains such as pypi.org and deb.debian.org for basic tool installation) to prevent the agent from cheating. 安全、内部子项或跨测试聚合仅保留原始披露,不进入常规展示。