GLM-5.2vsClaude Fable 5
Across 10 shared benchmarks, Claude Fable 5 leads overall: GLM-5.2 wins 0, Claude Fable 5 wins 10, with 0 ties and an average score difference of -19.41.
GLM-5.2
智谱AI · 2026-06-13 · Reasoning model
Claude Fable 5
Anthropic · 2026-06-09 · Reasoning model
GLM-5.20 wins(0%)(100%)10 winsClaude Fable 5
Benchmark scores
Grouped by capability, sorted by largest gap within each. 10 shared benchmarks.
Coding and Software Engineer
Claude Fable 5 3/3| Benchmark | GLM-5.2 | Claude Fable 5 | Diff |
|---|---|---|---|
| DeepSWE | 4421 / 27Deep Thinking (With Tools) | 702 / 27Deep Thinking (With Tools) | -26 |
| WeirdML v2 | 67.3121 / 52Thinking High (With Tools) | 87.853 / 52Thinking High (With Tools) | -20.54 |
| SWE-Bench Pro - Public | 62.109 / 57Thinking (With Tools) | 80.301 / 57Deep Thinking (With Tools) | -18.20 |
AI Agent - Tool Usage
Claude Fable 5 2/2| Benchmark | GLM-5.2 | Claude Fable 5 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 8117 / 44Thinking High (With Tools) | 884 / 44Deep Thinking (With Tools) | -7 |
| MCP-Atlas | 76.8013 / 38Thinking (With Tools) | 83.305 / 38Normal (With Tools) | -6.50 |
General Knowledge
Claude Fable 5 2/2| Benchmark | GLM-5.2 | Claude Fable 5 | Diff |
|---|---|---|---|
| HLE | 54.7015 / 181Thinking (With Tools) | 595 / 181Deep Thinking (No Tools) | -4.30 |
| LiveBench | 76.249 / 115Normal (No Tools) | 78.315 / 115Deep Thinking (No Tools) | -2.07 |
Math and Reasoning
Claude Fable 5 2/2| Benchmark | GLM-5.2 | Claude Fable 5 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 29.2716 / 34最高（无工具） | 87.801 / 34最高（无工具） | -58.54 |
| FrontierMath v2 | 59.2119 / 34最高（无工具） | 87.023 / 34最高（无工具） | -27.81 |
Commonsense Reasoning
Claude Fable 5 1/1| Benchmark | GLM-5.2 | Claude Fable 5 | Diff |
|---|---|---|---|
| SimpleBench | 58.8020 / 67Normal (No Tools) | 81.901 / 67Normal (No Tools) | -23.10 |
Specs
| Field | GLM-5.2 | Claude Fable 5 |
|---|---|---|
| Publisher | 智谱AI | Anthropic |
| Release date | 2026-06-13 | 2026-06-09 |
| Model type | Reasoning model | Reasoning model |
| Architecture | MoE | Dense |
| Parameters | 753.33B | Not available |
| Context length | 1M | 1M |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GLM-5.2 | Claude Fable 5 |
|---|---|---|
| Text input | $1.4 / 1M tokens | $10 / 1M tokens |
| Text output | $4.4 / 1M tokens | $50 / 1M tokens |
| Cache read | $0.26 / 1M tokens | $1 / 1M tokens |
| Cache write | Not public | $12.5 / 1M tokens |
Summary
- Claude Fable 5leads in:Coding and Software Engineer (3/3), AI Agent - Tool Usage (2/2), General Knowledge (2/2), Math and Reasoning (2/2), Commonsense Reasoning (1/1)
On average across the 10 shared benchmarks, Claude Fable 5 scores 19.41 higher.
Largest single-benchmark gap: FrontierMath Tier 4 v2 — GLM-5.2 29.27 vs Claude Fable 5 87.80 (-58.54).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.