GLM-5.3 abliterated (Anthropic)
By Anthropic, from GLM-5.3 by Z.ai
- Made by
- Anthropic
- Based on
- GLM-5.3 by Z.ai
- Released
- Announced in
- GLM-5.3: Frontier Coding with Emergent Cyber Capabilitiesz.ai
- Event
- GLM-5.3
- API record
- /api/v1/models/glm-5-3-abliterated-anthropic
Benchmark results
| Benchmark | Score | Source check |
|---|---|---|
| GPQA Diamond198 questions x 4 samples | 88% | CheckedA person checked this number against its source on 25 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
| CyberGym185-task subset | 81% | CheckedA person checked this number against its source on 25 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. |
Sources
Reported by Anthropic
- GLM-5.3 and the spread of advanced cyber capabilitiesanthropic.com · 29 Sep 2026
Original image
