GPT-5.6 SolvsGLM-5.2
Across 8 shared benchmarks, GPT-5.6 Sol leads overall: GPT-5.6 Sol wins 8, GLM-5.2 wins 0, with 0 ties and an average score difference of +21.59.
GPT-5.6 Sol
OpenAI · 2026-06-26 · Reasoning model
GLM-5.2
智谱AI · 2026-06-13 · Reasoning model
GPT-5.6 Sol8 wins(100%)(0%)0 winsGLM-5.2
Benchmark scores
Grouped by capability, sorted by largest gap within each. 8 shared benchmarks.
Coding and Software Engineer
GPT-5.6 Sol 4/4| Benchmark | GPT-5.6 Sol | GLM-5.2 | Diff |
|---|---|---|---|
| DeepSWE | 72.701 / 27极高强度思考（工具） | 4421 / 27Deep Thinking (With Tools) | +28.70 |
| Text Arena (Coding) | 1,6204 / 35极高强度思考（无工具） | 1,5935 / 35最高（无工具） | +27.02 |
| WeirdML v2 | 88.762 / 52Thinking High (With Tools) | 67.3121 / 52Thinking High (With Tools) | +21.45 |
| SWE-Bench Pro - Public | 64.607 / 57极高强度思考（工具） | 62.109 / 57Thinking (With Tools) | +2.50 |
Math and Reasoning
GPT-5.6 Sol 2/2| Benchmark | GPT-5.6 Sol | GLM-5.2 | Diff |
|---|---|---|---|
| FrontierMath Tier 4 v2 | 82.932 / 34最高（无工具） | 29.2716 / 34最高（无工具） | +53.66 |
| FrontierMath v2 | 89.121 / 34最高（无工具） | 59.2119 / 34最高（无工具） | +29.92 |
AI Agent - Tool Usage
GPT-5.6 Sol 1/1| Benchmark | GPT-5.6 Sol | GLM-5.2 | Diff |
|---|---|---|---|
| Terminal-Bench 2.1 | 88.801 / 44最高（无工具） | 8117 / 44Thinking High (With Tools) | +7.80 |
General Evaluation
GPT-5.6 Sol 1/1| Benchmark | GPT-5.6 Sol | GLM-5.2 | Diff |
|---|---|---|---|
| GPQA Diamond | 93.5014 / 226最高（无工具） | 91.8626 / 226最高（无工具） | +1.64 |
Specs
| Field | GPT-5.6 Sol | GLM-5.2 |
|---|---|---|
| Publisher | OpenAI | 智谱AI |
| Release date | 2026-06-26 | 2026-06-13 |
| Model type | Reasoning model | Reasoning model |
| Architecture | Dense | MoE |
| Parameters | Not available | 753.33B |
| Context length | 1.05M | 1M |
| Max output | 128K | 128K |
API pricing
Prices use DataLearner records when available; missing fields are not inferred.
| Item | GPT-5.6 Sol | GLM-5.2 |
|---|---|---|
| Text input | $5 / 1M tokens | $1.4 / 1M tokens |
| Text output | $30 / 1M tokens | $4.4 / 1M tokens |
| Cache read | $0.5 / 1M tokens | $0.26 / 1M tokens |
| Cache write | $6.25 / 1M tokens | Not public |
Summary
- GPT-5.6 Solleads in:Coding and Software Engineer (4/4), Math and Reasoning (2/2), AI Agent - Tool Usage (1/1), General Evaluation (1/1)
On average across the 8 shared benchmarks, GPT-5.6 Sol scores 21.59 higher.
Largest single-benchmark gap: FrontierMath Tier 4 v2 — GPT-5.6 Sol 82.93 vs GLM-5.2 29.27 (+53.66).
Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.