See key specs and per-benchmark scores for each model/mode. Scroll horizontally for all columns. 当前对比 2 个模型的评测数据与核心参数。
GLM-5
智谱AI
GLM-4.5
智谱AI
Best overall
GLM-5 · 71.40
Best single
GLM-5 · GPQA Diamond 86.00
Modality coverage
GLM-5 · 1 modalities
Head to head
3
Benchmarks
3
Wins
0
Losses
+18.83
Average diff
Compare benchmark results across thinking modes and tool usage.
Data sourced primarily from official releases (GitHub, Hugging Face, papers), then benchmark leaderboards, then third-party evaluators. Learn about our data methodology
Complete scores for each model/mode across selected benchmarks.
3 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | GLM-5 | GLM-4.5 |
|---|---|---|
GPQA Diamond 综合评估 | 86.00Thinking Enabled | 79.10Thinking Enabled |
HLE 综合评估 | 50.40Thinking Enabled ｜ Tools | 14.40Thinking Enabled |
SWE-bench Verified 编程与软件工程 | 77.80Thinking Enabled | 64.20Thinking Enabled |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GLM-5智谱AI | GLM-4.5智谱AI |
|---|---|---|
Core specsRelease | 2026-02-11 | 2025-07-28 |
Context length | 200K | 128K |
Parameters | 7440 | 3550 |
Active parameters | 400 | 320 |
Max output | 131072 | 97280 |
MoE | Yes | Yes |
Supported modes | No mode data | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode） |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Closed Source | Not provided |
Commercial use | 免费商用授权 | 免费商用授权 |
Modality supportText Input/Output | / | / |
ResourcesPaper / report | GLM-5: From Vibe Coding to Agentic Engineering | GLM-4.5: Reasoning, Coding, and Agentic Abililties |
DataLearner blog | Not provided | Zhipu AI重磅发布GLM-4.5系列：技术深度解析与多维度性能评测 |
