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。

DeepSeek-V4-Flash
DeepSeek-AI

GLM 5.1
智谱AI
Each axis is a category average, normalized to a 100-point radar.
Relative edge: AI Agent - 工具使用 +7.4 / Relative gap: AI Agent - 信息收集 -15.9
Relative edge: AI Agent - 信息收集 +15.9 / Relative gap: AI Agent - 工具使用 -7.4
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
DeepSeek-V4-Flash · 201.90
Best single
DeepSeek-V4-Flash · Text Arena (Coding) 1576.54
Modality coverage
DeepSeek-V4-Flash · 1 modalities
Head to head
11
Benchmarks
4
Wins
7
Losses
+1.70
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 | DeepSeek-V4-Flash | GLM 5.1 |
|---|---|---|
HLE 综合评估 | 45.10Thinking Level · Extra High | Tools | 52.30Thinking Enabled | Tools |
LiveBench 综合评估 | 67.25Standard Mode | 70.18Standard Mode |
GPQA Diamond 科学与综合推理 | 88.10Thinking Level · High | 86.20Thinking Enabled |
SWE-Bench Pro - Public 编程与软件工程 | 52.60Thinking Level · Extra High | Tools | 58.40Thinking Enabled | Tools |
Text Arena (Coding) 编程与软件工程 | 1576.54Thinking Level · High | 1534.00Standard Mode |
WeirdML v2 编程与软件工程 | 43.76Thinking Level · High | Tools | 57.10Standard Mode | Tools |
SimpleBench 常识推理 | 46.30Standard Mode | 58.70Standard Mode |
BrowseComp AI Agent - 信息收集 | 73.20Thinking Level · Extra High | Tools | 79.30Thinking Enabled | Tools |
Terminal Bench 2.0 AI Agent - 工具使用 | 56.90Thinking Level · Extra High | Tools | 63.50Thinking Enabled | Tools |
Terminal-Bench 2.1 AI Agent - 工具使用 | 82.70Thinking Level · High | Tools | 58.70Thinking Level · High | Tools |
IMO-AnswerBench 数学推理 | 88.40Thinking Level · High | 83.80Thinking Enabled |
Side-by-side input/output token pricing
Licensing, MoE architecture, and multi-modality support.
| Features & specs | DeepSeek-V4-FlashDeepSeek-AI | GLM 5.1智谱AI |
|---|---|---|
Core specsRelease | 2026-04-24 | 2026-03-27 |
Context length | 1M | 200K |
Parameters | 2840 | 754 |
Active parameters | 130 | 40 |
Max output | 384000 | 128000 |
MoE | Yes | Yes |
LicenseCode Open Source | Closed Source | Closed Source |
Weights Open Source | Closed Source | Closed Source |
Commercial use | 免费商用授权 | 免费商用授权 |
Modality supportText Input/Output | / | / |
ResourcesPaper / report | DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence | GLM-5.1: Towards Long-Horizon Tasks |