Compare · ModelsLive · 2 picked · head to head
GLM 5.2 vs Qwen3.7 Max
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GLM 5.2 wins on 15/24 benchmarks
GLM 5.2 wins 15 of 24 shared benchmarks. Leads in speed · knowledge · general.
Category leads
speed·GLM 5.2knowledge·GLM 5.2general·GLM 5.2math·Qwen3.7 Maxcoding·GLM 5.2reasoning·Qwen3.7 Maxlanguage·Qwen3.7 Max
Hype vs Reality
Attention vs performance
GLM 5.2
#67 by perf·#3 by attention
Qwen3.7 Max
#41 by perf·#2 by attention
Best value
Qwen3.7 Max
1.1x better value than GLM 5.2
GLM 5.2
22.6 pts/$
$2.50/M
Qwen3.7 Max
24.5 pts/$
$2.50/M
Vendor risk
Who is behind the model
z-ai
private · undisclosed
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
24 benchmarks · 2 models
GLM 5.2Qwen3.7 Max
Artificial Analysis · Agentic Index
GLM 5.2 leads by +12.5
Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?"
GLM 5.2
43.1
Qwen3.7 Max
30.6
Artificial Analysis · Coding Index
GLM 5.2 leads by +2.8
Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads.
GLM 5.2
68.8
Qwen3.7 Max
66.0
Artificial Analysis · Quality Index
GLM 5.2 leads by +5.1
GLM 5.2
51.1
Qwen3.7 Max
46.0
Chess Puzzles
GLM 5.2 leads by +2.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GLM 5.2
16.9
Qwen3.7 Max
14.8
Dtbench
GLM 5.2 leads by +2.2
GLM 5.2
89.3
Qwen3.7 Max
87.1
Ebr Bench
GLM 5.2
9.5
Qwen3.7 Max
9.5
FrontierMath-Tier-4-v2-Private
Qwen3.7 Max leads by +4.9
GLM 5.2
29.3
Qwen3.7 Max
34.1
FrontierMath-Tiers-1-3-v2-Private
Qwen3.7 Max leads by +5.4
GLM 5.2
59.2
Qwen3.7 Max
64.6
GPQA diamond
GLM 5.2 leads by +1.3
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GLM 5.2
89.1
Qwen3.7 Max
87.9
LiveBench · Agentic Coding
GLM 5.2 leads by +21.7
GLM 5.2
73.3
Qwen3.7 Max
51.7
LiveBench · Coding
GLM 5.2 leads by +5.4
GLM 5.2
79.7
Qwen3.7 Max
74.2
LiveBench · Data Analysis
GLM 5.2 leads by +1.9
GLM 5.2
73.7
Qwen3.7 Max
71.8
LiveBench · If
Qwen3.7 Max leads by +11.8
GLM 5.2
62.3
Qwen3.7 Max
74.0
LiveBench · Language
Qwen3.7 Max leads by +3.5
GLM 5.2
76.2
Qwen3.7 Max
79.7
LiveBench · Mathematics
GLM 5.2 leads by +4.5
GLM 5.2
89.8
Qwen3.7 Max
85.3
LiveBench · Overall
GLM 5.2 leads by +1.9
GLM 5.2
76.2
Qwen3.7 Max
74.3
LiveBench · Reasoning
Qwen3.7 Max leads by +4.7
GLM 5.2
78.6
Qwen3.7 Max
83.3
Lmca
GLM 5.2 leads by +2.1
GLM 5.2
53.9
Qwen3.7 Max
51.8
Mystery Game Puzzles
Qwen3.7 Max leads by +14.3
GLM 5.2
10.8
Qwen3.7 Max
25.1
OTIS Mock AIME 2024-2025
Qwen3.7 Max leads by +9.2
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GLM 5.2
86.4
Qwen3.7 Max
95.5
Proofbench
GLM 5.2 leads by +9.0
GLM 5.2
35.0
Qwen3.7 Max
26.0
SimpleBench
Qwen3.7 Max leads by +13.9
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
GLM 5.2
50.6
Qwen3.7 Max
64.5
SimpleQA Verified
Qwen3.7 Max leads by +21.6
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GLM 5.2
34.2
Qwen3.7 Max
55.8
SWE-Bench verified
GLM 5.2 leads by +1.4
SWE-bench Verified · 500 human-validated tasks from 12 real Python repositories (Django, Flask, scikit-learn, sympy, and others). Each task requires the model to produce a git patch that resolves a real GitHub issue and passes the test suite. The verified subset eliminates ambiguous tasks from the original SWE-bench. Claude Mythos Preview leads at 93.9%, crossing 90% for the first time in 2026. Opus 4.6 scores 80.8%. The benchmark remains the most-cited evaluation for code-generation capability.
GLM 5.2
78.7
Qwen3.7 Max
77.3
Full benchmark table
| Benchmark | GLM 5.2 | Qwen3.7 Max |
|---|---|---|
Artificial Analysis · Agentic Index Artificial Analysis Agentic Index · a composite score measuring how well a model performs in agentic workflows · multi-step tool use, planning, error recovery, and autonomous task completion. Aggregates results from multiple agentic benchmarks including SWE-bench, tool-use tests, and planning evaluations. The canonical single-number metric for "how good is this model as an agent?" | 43.1 | 30.6 |
Artificial Analysis · Coding Index Artificial Analysis Coding Index · a composite score that aggregates performance across multiple coding benchmarks into a single index. Tracks code generation quality, debugging ability, multi-language competence, and real-world software engineering tasks. Used by Artificial Analysis to rank model coding capability in a normalized, comparable format. Useful for developers choosing between models for coding-heavy workloads. | 68.8 | 66.0 |
Artificial Analysis · Quality Index | 51.1 | 46.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 16.9 | 14.8 |
Dtbench | 89.3 | 87.1 |
Ebr Bench | 9.5 | 9.5 |
FrontierMath-Tier-4-v2-Private | 29.3 | 34.1 |
FrontierMath-Tiers-1-3-v2-Private | 59.2 | 64.6 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 89.1 | 87.9 |
LiveBench · Agentic Coding | 73.3 | 51.7 |
LiveBench · Coding | 79.7 | 74.2 |
LiveBench · Data Analysis | 73.7 | 71.8 |
LiveBench · If | 62.3 | 74.0 |
LiveBench · Language | 76.2 | 79.7 |
LiveBench · Mathematics | 89.8 | 85.3 |
LiveBench · Overall | 76.2 | 74.3 |
LiveBench · Reasoning | 78.6 | 83.3 |
Lmca | 53.9 | 51.8 |
Mystery Game Puzzles | 10.8 | 25.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 86.4 | 95.5 |
Proofbench | 35.0 | 26.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 50.6 | 64.5 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 34.2 | 55.8 |
SWE-Bench verified SWE-bench Verified · 500 human-validated tasks from 12 real Python repositories (Django, Flask, scikit-learn, sympy, and others). Each task requires the model to produce a git patch that resolves a real GitHub issue and passes the test suite. The verified subset eliminates ambiguous tasks from the original SWE-bench. Claude Mythos Preview leads at 93.9%, crossing 90% for the first time in 2026. Opus 4.6 scores 80.8%. The benchmark remains the most-cited evaluation for code-generation capability. | 78.7 | 77.3 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.00 | $4.00 | 1.0M tokens (~524 books) | $17.50 | |
| $1.25 | $3.75 | 1.0M tokens (~500 books) | $18.75 |