GLM 5.2
ModelOpen sourceActiveby Z.ai (Zhipu AI) · family “glm”
GLM 5.2 is a large-scale reasoning model from Z.ai. It supports text input and output with a 1M-token context window, and is suited for long-horizon agent workflows, project-level software engineering,... (description from the OpenRouter listing)
in: textout: textReasoningTool useStructured output
Epoch Capabilities Index
151.8 #43 of 268
90% CI 149.9 – 153.6
Released
16 Jun 2026
Context window
1.05M tokens
Max output
944K tokens
Input $ / 1M tokens
$0.39
Output $ / 1M tokens
$4.40
Cached input $ / 1M
$0.26
Each benchmark shown separately with its own source
Benchmark results (20)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 45.8% | 67% | max | — | External ↗ |
| DTBench | reasoning | 93.6% | 95% | max | — | External ↗ |
| Mystery Game Puzzles | games | 19.0% ±3.9 | 23% | low | 27 Aug 2026 | Epoch ↗ |
| EBR-bench | reasoning | 9.5% | 13% | max | 29 Jun 2026 | Epoch ↗ |
| Surface Evolver Bench | science | 55.6% | 59% | high | — | External ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 59.2% ±3.0 | 63% | max | 19 Jun 2026 | Epoch ↗ |
| FrontierMath-Tier-4-v2-Private | math | 29.3% ±7.2 | 30% | max | 19 Jun 2026 | Eval log ↗ |
| FrontierCode | coding | 24.5% | 46% | none | — | External ↗ |
| DeepSWE | coding | 43.8% | 59% | max | — | External ↗ |
| PostTrainBench | agents | 31.7% | 76% | max | — | External ↗ |
| ProofBench | math | 35.0% | 35% | max | — | External ↗ |
| Chess Puzzles | games | 21.0% ±4.1 | 29% | max | 17 Jun 2026 | Epoch ↗ |
| SimpleQA Verified | knowledge | 34.2% ±1.5 | 45% | max | 27 Aug 2026 | Eval log ↗ |
| ARC-AGI-2 | reasoning | 22.8% | 24% | unknown | — | External ↗ |
| WeirdML | coding | 70.1% ±0.0 | 75% | max | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 86.4% ±4.2 | 86% | max | 25 Jun 2026 | Epoch ↗ |
| SimpleBench | reasoning | 58.8% | 72% | unknown | — | External ↗ |
| SWE-Bench verified | coding | 78.7% ±1.9 | 94% | max | 25 Jun 2026 | Eval log ↗ |
| GPQA diamond | science | 91.9% ±1.6 | 96% | max | 24 Jun 2026 | Epoch ↗ |
| ARC-AGI | reasoning | 77.0% | 78% | unknown | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
6 recorded prices on 17 Jul 2026 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 7h ago. Steps show when the price changed.
Events
- Price cut-15%GLM 5.2 API price cutOpenRouter models API
- BenchmarkEpoch AI evaluates GLM 5.2Epoch AI Benchmarking Hub
- BenchmarkEpoch AI evaluates GLM 5.2Epoch AI Benchmarking Hub
- Open releaseZ.ai releases GLM 5.2Epoch AI Benchmarking Hub