Compare · ModelsLive · 2 picked · head to head
GPT-5 Mini vs Qwen3.6 Plus
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Qwen3.6 Plus wins on 13/19 benchmarks
Qwen3.6 Plus wins 13 of 19 shared benchmarks. Leads in knowledge · general · math.
Category leads
knowledge·Qwen3.6 Plusgeneral·Qwen3.6 Plusmath·Qwen3.6 Pluscoding·Qwen3.6 Plusreasoning·Qwen3.6 Pluslanguage·GPT-5 Mini
Hype vs Reality
Attention vs performance
GPT-5 Mini
#108 by perf·#4 by attention
Qwen3.6 Plus
#94 by perf·#2 by attention
Best value
Qwen3.6 Plus
1.0x better value than GPT-5 Mini
GPT-5 Mini
45.8 pts/$
$1.13/M
Qwen3.6 Plus
47.0 pts/$
$1.14/M
Vendor risk
Who is behind the model
OpenAI
$840.0B·Tier 1
Alibaba (Qwen)
$293.0B·Tier 1
Head to head
19 benchmarks · 2 models
GPT-5 MiniQwen3.6 Plus
Chess Puzzles
GPT-5 Mini leads by +13.7
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
GPT-5 Mini
26.4
Qwen3.6 Plus
12.7
Dtbench
Qwen3.6 Plus leads by +2.2
GPT-5 Mini
67.5
Qwen3.6 Plus
69.8
FrontierMath-2025-02-28-Private
GPT-5 Mini leads by +1.0
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
GPT-5 Mini
27.2
Qwen3.6 Plus
26.2
FrontierMath-Tier-4-2025-07-01-Private
Qwen3.6 Plus leads by +7.6
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
GPT-5 Mini
6.3
Qwen3.6 Plus
13.9
FrontierMath-Tiers-1-3-v2-Private
GPT-5 Mini leads by +8.4
GPT-5 Mini
46.7
Qwen3.6 Plus
38.3
GPQA diamond
Qwen3.6 Plus leads by +17.8
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
GPT-5 Mini
66.7
Qwen3.6 Plus
84.5
LiveBench · Agentic Coding
Qwen3.6 Plus leads by +20.0
GPT-5 Mini
35.0
Qwen3.6 Plus
55.0
LiveBench · Coding
Qwen3.6 Plus leads by +2.1
GPT-5 Mini
76.1
Qwen3.6 Plus
78.2
LiveBench · Data Analysis
Qwen3.6 Plus leads by +20.3
GPT-5 Mini
49.6
Qwen3.6 Plus
69.9
LiveBench · If
GPT-5 Mini leads by +5.9
GPT-5 Mini
64.2
Qwen3.6 Plus
58.3
LiveBench · Language
Qwen3.6 Plus leads by +5.8
GPT-5 Mini
69.2
Qwen3.6 Plus
75.0
LiveBench · Mathematics
Qwen3.6 Plus leads by +9.3
GPT-5 Mini
74.4
Qwen3.6 Plus
83.7
LiveBench · Overall
Qwen3.6 Plus leads by +9.8
GPT-5 Mini
61.0
Qwen3.6 Plus
70.8
LiveBench · Reasoning
Qwen3.6 Plus leads by +17.2
GPT-5 Mini
58.6
Qwen3.6 Plus
75.8
Lmca
GPT-5 Mini leads by +1.3
GPT-5 Mini
40.2
Qwen3.6 Plus
38.9
Mystery Game Puzzles
Qwen3.6 Plus leads by +2.2
GPT-5 Mini
0.9
Qwen3.6 Plus
3.1
OTIS Mock AIME 2024-2025
Qwen3.6 Plus leads by +6.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
GPT-5 Mini
86.7
Qwen3.6 Plus
93.3
SimpleQA Verified
Qwen3.6 Plus leads by +22.5
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
GPT-5 Mini
21.6
Qwen3.6 Plus
44.1
SWE-Bench verified
GPT-5 Mini leads by +6.8
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.
GPT-5 Mini
64.7
Qwen3.6 Plus
57.9
Full benchmark table
| Benchmark | GPT-5 Mini | Qwen3.6 Plus |
|---|---|---|
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 26.4 | 12.7 |
Dtbench | 67.5 | 69.8 |
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. | 27.2 | 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. | 6.3 | 13.9 |
FrontierMath-Tiers-1-3-v2-Private | 46.7 | 38.3 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 66.7 | 84.5 |
LiveBench · Agentic Coding | 35.0 | 55.0 |
LiveBench · Coding | 76.1 | 78.2 |
LiveBench · Data Analysis | 49.6 | 69.9 |
LiveBench · If | 64.2 | 58.3 |
LiveBench · Language | 69.2 | 75.0 |
LiveBench · Mathematics | 74.4 | 83.7 |
LiveBench · Overall | 61.0 | 70.8 |
LiveBench · Reasoning | 58.6 | 75.8 |
Lmca | 40.2 | 38.9 |
Mystery Game Puzzles | 0.9 | 3.1 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 86.7 | 93.3 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 21.6 | 44.1 |
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. | 64.7 | 57.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $0.25 | $2.00 | 400K tokens (~200 books) | $6.88 | |
| $0.33 | $1.95 | 1.0M tokens (~500 books) | $7.31 |