Compare · ModelsLive · 2 picked · head to head
Gemini 3.5 Flash vs Claude Opus 4.6
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Gemini 3.5 Flash wins on 9/16 benchmarks
Gemini 3.5 Flash wins 9 of 16 shared benchmarks. Leads in agentic · reasoning · knowledge.
Category leads
agentic·Gemini 3.5 Flashreasoning·Gemini 3.5 Flasharena·Claude Opus 4.6knowledge·Gemini 3.5 Flashmath·Claude Opus 4.6coding·Gemini 3.5 Flash
Hype vs Reality
Attention vs performance
Gemini 3.5 Flash
#44 by perf·no signal
Claude Opus 4.6
#72 by perf·#4 by attention
Best value
Gemini 3.5 Flash
3.1x better value than Claude Opus 4.6
Gemini 3.5 Flash
11.9 pts/$
$5.25/M
Claude Opus 4.6
3.8 pts/$
$15.00/M
Vendor risk
Who is behind the model
Google DeepMind
$4.00T·Tier 1
Anthropic
$380.0B·Tier 1
Head to head
16 benchmarks · 2 models
Gemini 3.5 FlashClaude Opus 4.6
APEX-Agents
Gemini 3.5 Flash leads by +17.9
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 3.5 Flash
49.6
Claude Opus 4.6
31.7
ARC-AGI
Claude Opus 4.6 leads by +1.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Gemini 3.5 Flash
92.5
Claude Opus 4.6
94.0
ARC-AGI-2
Gemini 3.5 Flash leads by +2.9
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Gemini 3.5 Flash
72.1
Claude Opus 4.6
69.2
Chatbot Arena Elo · Coding
Claude Opus 4.6 leads by +32.7
Gemini 3.5 Flash
1505.5
Claude Opus 4.6
1538.3
Chatbot Arena Elo · Overall
Claude Opus 4.6 leads by +22.5
Gemini 3.5 Flash
1476.3
Claude Opus 4.6
1498.8
Chess Puzzles
Gemini 3.5 Flash leads by +33.0
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 3.5 Flash
50.0
Claude Opus 4.6
17.0
FrontierMath-2025-02-28-Private
Claude Opus 4.6 leads by +1.7
FrontierMath (Feb 2025) · original research-level math problems created by mathematicians, testing capabilities at the boundary of current AI mathematical reasoning.
Gemini 3.5 Flash
39.0
Claude Opus 4.6
40.7
FrontierMath-Tier-4-2025-07-01-Private
Claude Opus 4.6 leads by +8.3
FrontierMath Tier 4 (Jul 2025) · the most challenging tier of frontier mathematics, containing problems that push the absolute limits of AI mathematical reasoning.
Gemini 3.5 Flash
14.6
Claude Opus 4.6
22.9
FrontierMath-Tier-4-v2-Private
Gemini 3.5 Flash
26.8
Claude Opus 4.6
26.8
FrontierMath-Tiers-1-3-v2-Private
Claude Opus 4.6 leads by +3.1
Gemini 3.5 Flash
62.8
Claude Opus 4.6
66.0
GPQA diamond
Gemini 3.5 Flash leads by +3.0
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 3.5 Flash
90.4
Claude Opus 4.6
87.4
OTIS Mock AIME 2024-2025
Gemini 3.5 Flash leads by +1.1
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 3.5 Flash
95.5
Claude Opus 4.6
94.4
SimpleBench
Gemini 3.5 Flash leads by +10.9
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Gemini 3.5 Flash
72.0
Claude Opus 4.6
61.1
SimpleQA Verified
Gemini 3.5 Flash leads by +21.9
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 3.5 Flash
68.4
Claude Opus 4.6
46.5
SWE-Bench verified
Gemini 3.5 Flash leads by +0.6
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.
Gemini 3.5 Flash
79.3
Claude Opus 4.6
78.7
WeirdML
Claude Opus 4.6 leads by +15.3
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Gemini 3.5 Flash
62.6
Claude Opus 4.6
78.0
Full benchmark table
| Benchmark | Gemini 3.5 Flash | Claude Opus 4.6 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 49.6 | 31.7 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 92.5 | 94.0 |
ARC-AGI-2 ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data. | 72.1 | 69.2 |
Chatbot Arena Elo · Coding | 1505.5 | 1538.3 |
Chatbot Arena Elo · Overall | 1476.3 | 1498.8 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 50.0 | 17.0 |
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. | 39.0 | 40.7 |
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. | 14.6 | 22.9 |
FrontierMath-Tier-4-v2-Private | 26.8 | 26.8 |
FrontierMath-Tiers-1-3-v2-Private | 62.8 | 66.0 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 90.4 | 87.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.5 | 94.4 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 72.0 | 61.1 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 68.4 | 46.5 |
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. | 79.3 | 78.7 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 62.6 | 78.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $1.50 | $9.00 | 1.0M tokens (~524 books) | $33.75 | |
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 |