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-4.7
智谱AI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: none clear / Relative gap: 编程与软件工程 -21.5
Relative edge: 编程与软件工程 +21.5 / Relative gap: none clear
Method: for each model and benchmark, the chart first averages all scores in the current mode scope instead of taking the best score, then averages those benchmark scores within each category. Only benchmarks with at least two selected models scored are included; missing values are not counted as zero.
Best overall
GLM-5.2 · 74.24
Best single
GLM-5.2 · AIME 2026 99.20
Modality coverage
GLM-4.7 · 1 modalities
Head to head
7
Benchmarks
0
Wins
7
Losses
-13.40
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.
7 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | GLM-4.7 | GLM-5.2 |
|---|---|---|
HLE 综合评估 | 42.80Thinking Enabled ｜ Tools | 54.70Thinking Enabled ｜ Tools |
LiveBench 综合评估 | 58.09Standard Mode | 76.24Standard Mode |
GPQA Diamond 科学与综合推理 | 85.70Thinking Enabled | 91.86Thinking Level · High |
SWE-Bench Pro - Public 编程与软件工程 | 40.60Thinking Enabled ｜ Tools | 62.10Thinking Enabled ｜ Tools |
AIME 2026 数学推理 | 92.90Thinking Enabled | 99.20Thinking Enabled |
SimpleBench 常识推理 | 47.70Thinking Enabled | 58.80Standard Mode |
MCP-Atlas AI Agent - 工具使用 | 58.10Standard Mode ｜ Tools | 76.80Thinking Enabled ｜ Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GLM-4.7智谱AI | GLM-5.2智谱AI |
|---|---|---|
Core specsRelease | 2025-12-22 | 2026-06-13 |
Context length | 200K | 1M |
Parameters | 358B | 753.3B |
Active parameters | 32B | 40B |
Max output | 132072 | 128000 |
MoE | Yes | Yes |
Supported modes | 常规模式（Non-Thinking Mode）思考模式（Thinking Mode） | No mode data |
LicenseCode Open Source | Open Source · MIT License | Open Source · MIT License |
Weights Open Source | Open Source · MIT License | Open Source · MIT License |
Commercial use | 免费商用授权 | 免费商用授权 |
Local deploymentWeight size | Not provided | 约 1.5 TB（BF16 safetensors） |
VRAM for weights | Not provided | ≈ 1500 GB (weights only, excludes KV cache) |
Weights | Hugging Face | Hugging Face |
Source repo | GitHub | GitHub |
Modality supportText Input/Output | / | / |
ResourcesPaper / report | GLM-4.7: Advancing the Coding Capability | GLM-5: from Vibe Coding to Agentic Engineering |
GLM-5.2
智谱AI
