Compare · ModelsLive · 3 picked · head to head
GPT-5.2-Codex vs GLM 5.1 vs Qwen3.6 Plus
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GLM 5.1 wins on 12/20 benchmarks
GLM 5.1 wins 12 of 20 shared benchmarks. Leads in coding · language · math.
Category leads
coding·GLM 5.1reasoning·GPT-5.2-Codexlanguage·GLM 5.1math·GLM 5.1knowledge·Qwen3.6 Plusspeed·GLM 5.1arena·GLM 5.1
Hype vs Reality
Attention vs performance
GPT-5.2-Codex
#18 by perf·no signal
GLM 5.1
#56 by perf·no signal
Qwen3.6 Plus
#55 by perf·no signal
Best value
Qwen3.6 Plus
1.8x better value than GLM 5.1
GPT-5.2-Codex
9.0 pts/$
$7.88/M
GLM 5.1
29.9 pts/$
$2.00/M
Qwen3.6 Plus
52.7 pts/$
$1.14/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
z-ai
private · undisclosed
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
20 benchmarks · 3 models
GPT-5.2-CodexGLM 5.1Qwen3.6 Plus
LiveBench · Agentic Coding
GPT-5.2-Codex
51.7
GLM 5.1
55.0
Qwen3.6 Plus
55.0
LiveBench · Coding
GPT-5.2-Codex leads by +5.4
GPT-5.2-Codex
83.6
GLM 5.1
75.4
Qwen3.6 Plus
78.2
LiveBench · Data Analysis
GPT-5.2-Codex leads by +8.3
GPT-5.2-Codex
78.2
GLM 5.1
63.2
Qwen3.6 Plus
69.9
LiveBench · If
GLM 5.1 leads by +2.0
GPT-5.2-Codex
66.5
GLM 5.1
68.5
Qwen3.6 Plus
58.3
LiveBench · Language
Qwen3.6 Plus leads by +1.3
GPT-5.2-Codex
73.7
GLM 5.1
71.8
Qwen3.6 Plus
75.0
LiveBench · Mathematics
GPT-5.2-Codex leads by +3.9
GPT-5.2-Codex
88.8
GLM 5.1
84.9
Qwen3.6 Plus
83.7
LiveBench · Overall
GPT-5.2-Codex leads by +3.5
GPT-5.2-Codex
74.3
GLM 5.1
70.2
Qwen3.6 Plus
70.8
LiveBench · Reasoning
GPT-5.2-Codex leads by +1.9
GPT-5.2-Codex
77.7
GLM 5.1
72.5
Qwen3.6 Plus
75.8
Artificial Analysis · Agentic Index
GLM 5.1 leads by +2.3
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 5.1
29.9
Qwen3.6 Plus
27.6
Artificial Analysis · Coding Index
GLM 5.1 leads by +1.3
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 5.1
55.8
Qwen3.6 Plus
54.5
Artificial Analysis · Quality Index
GLM 5.1 leads by +0.6
GLM 5.1
40.2
Qwen3.6 Plus
39.6
Chatbot Arena Elo · Coding
GLM 5.1 leads by +67.3
GLM 5.1
1529.2
Qwen3.6 Plus
1461.9
Chatbot Arena Elo · Overall
GLM 5.1 leads by +31.0
GLM 5.1
1475.3
Qwen3.6 Plus
1444.2
Chess Puzzles
GLM 5.1 leads by +1.8
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GLM 5.1
18.0
Qwen3.6 Plus
16.2
FrontierMath-2025-02-28-Private
GLM 5.1 leads by +7.2
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GLM 5.1
33.5
Qwen3.6 Plus
26.2
FrontierMath-Tier-4-2025-07-01-Private
GLM 5.1 leads by +4.2
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GLM 5.1
12.5
Qwen3.6 Plus
8.3
GPQA diamond
Qwen3.6 Plus leads by +2.5
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GLM 5.1
80.6
Qwen3.6 Plus
83.2
OTIS Mock AIME 2024-2025
GLM 5.1 leads by +1.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GLM 5.1
92.2
Qwen3.6 Plus
90.5
SimpleQA Verified
Qwen3.6 Plus leads by +11.8
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GLM 5.1
37.3
Qwen3.6 Plus
49.1
SWE-Bench verified
GLM 5.1 leads by +16.3
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 5.1
74.2
Qwen3.6 Plus
57.9
Full benchmark table
| Benchmark | GPT-5.2-Codex | GLM 5.1 | Qwen3.6 Plus |
|---|---|---|---|
LiveBench · Agentic Coding | 51.7 | 55.0 | 55.0 |
LiveBench · Coding | 83.6 | 75.4 | 78.2 |
LiveBench · Data Analysis | 78.2 | 63.2 | 69.9 |
LiveBench · If | 66.5 | 68.5 | 58.3 |
LiveBench · Language | 73.7 | 71.8 | 75.0 |
LiveBench · Mathematics | 88.8 | 84.9 | 83.7 |
LiveBench · Overall | 74.3 | 70.2 | 70.8 |
LiveBench · Reasoning | 77.7 | 72.5 | 75.8 |
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?" | — | 29.9 | 27.6 |
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. | — | 55.8 | 54.5 |
Artificial Analysis · Quality Index | — | 40.2 | 39.6 |
Chatbot Arena Elo · Coding | — | 1529.2 | 1461.9 |
Chatbot Arena Elo · Overall | — | 1475.3 | 1444.2 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | — | 18.0 | 16.2 |
FrontierMath-2025-02-28-Private FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning. | — | 33.5 | 26.2 |
FrontierMath-Tier-4-2025-07-01-Private FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning. | — | 12.5 | 8.3 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | — | 80.6 | 83.2 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | — | 92.2 | 90.5 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | — | 37.3 | 49.1 |
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. | — | 74.2 | 57.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.75 | $14.00 | 400K tokens (~200 books) | $48.13 | |
| $0.97 | $3.04 | 203K tokens (~101 books) | $14.83 | |
| $0.33 | $1.95 | 1.0M tokens (~500 books) | $7.31 |