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-5.3
智谱AI

Kimi K3
Moonshot AI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: 综合评估 +12.8 / Relative gap: 编程与软件工程 -11.8
Relative edge: 编程与软件工程 +11.8 / Relative gap: 综合评估 -12.8
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-5.3 · 210.52
Best single
GLM-5.3 · GDPval-AA v2 1769.00
Modality coverage
Kimi K3 · 3 modalities
Head to head
11
Benchmarks
6
Wins
5
Losses
+4.06
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.
11 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | GLM-5.3 | Kimi K3 |
|---|---|---|
HLE 综合评估 | 62.50Thinking Level · High | Tools | 56.00Thinking Level · High | Tools |
Automation Bench AI Agent - 工具使用 | 48.20Thinking Level · High | Tools | 30.80Thinking Level · High | Tools |
Terminal-Bench 2.1 AI Agent - 工具使用 | 88.20Thinking Level · High | Tools | 88.30Thinking Level · High | Tools |
Toolathlon-Verified AI Agent - 工具使用 | 73.00Thinking Level · High | Tools | 76.50Thinking Level · High | Tools |
DeepSWE 编程与软件工程 | 66.90Thinking Level · High | Tools | 67.50Thinking Level · High | Tools |
FrontierSWE 编程与软件工程 | 78.10Thinking Level · High | Tools | 81.20Thinking Level · High | Tools |
PostTrain Bench 编程与软件工程 | 39.80Thinking Level · High | Tools | 36.60Thinking Level · High | Tools |
Program Bench 编程与软件工程 | 19.00Thinking Level · High | Tools | 77.80Thinking Level · High | Tools |
SWE-Marathon 编程与软件工程 | 42.50Thinking Level · High | Tools | 42.00Thinking Level · High | Tools |
Agents' Last Exam Agent能力评测 | 28.50Thinking Level · High | Tools | 28.30Thinking Level · High | Tools |
GDPval-AA v2 生产力知识 | 1769.00Thinking Level · High | Tools | 1686.00Thinking Level · High | Tools |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | GLM-5.3智谱AI | Kimi K3Moonshot AI |
|---|---|---|
Core specsRelease | 2026-08-14 | 2026-07-16 |
Context length | 1M | 1M |
Parameters | 7533.3 | 28000 |
Active parameters | Not provided | 1040 |
Max output | 128000 | 1048576 |
MoE | Yes | Yes |
LicenseCode Open Source | Not provided | Not provided |
Weights Open Source | Not provided | Not provided |
Commercial use | Not provided | 有条件免费商用授权 |
Modality supportText Input/Output | / | / |
Image Input/Output | Not provided | / |
Video Input/Output | Not provided | / |
ResourcesPaper / report | GLM-5.3: Frontier Coding with Emergent Cyber Capabilities | Kimi K3 技术报告 |