Compare · ModelsLive · 3 picked · head to head
Gemini 4 Argon vs Gemini 4 Carbon vs GPT-6 Astra
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
GPT-6 Astra wins on 15/28 benchmarks
GPT-6 Astra wins 15 of 28 shared benchmarks. Leads in general · math · coding.
Category leads
agentic·Gemini 4 Carbonreasoning·Gemini 4 Argonknowledge·Gemini 4 Carbongeneral·GPT-6 Astramath·GPT-6 Astracoding·GPT-6 Astraspeed·GPT-6 Astra
Hype vs Reality
Attention vs performance
Gemini 4 Argon
#17 by perf·#8 by attention
Gemini 4 Carbon
#1 by perf·#8 by attention
GPT-6 Astra
#9 by perf·#2 by attention
Best value
Gemini 4 Carbon
1.2x better value than Gemini 4 Argon
Gemini 4 Argon
12.0 pts/$
$6.00/M
Gemini 4 Carbon
14.3 pts/$
$6.00/M
GPT-6 Astra
2.5 pts/$
$30.00/M
Vendor risk
Who is behind the model
Google DeepMind
$3.61T·Tier 1
Google DeepMind
$3.61T·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
28 benchmarks · 3 models
Gemini 4 ArgonGemini 4 CarbonGPT-6 Astra
APEX-Agents
Gemini 4 Carbon leads by +1.8
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 4 Argon
82.4
Gemini 4 Carbon
84.2
GPT-6 Astra
64.7
ARC-AGI
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Gemini 4 Argon
98.5
Gemini 4 Carbon
98.5
GPT-6 Astra
98.5
ARC-AGI-2
GPT-6 Astra leads by +0.8
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Gemini 4 Argon
91.3
Gemini 4 Carbon
94.2
GPT-6 Astra
95.0
Chess Puzzles
Gemini 4 Carbon leads by +2.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 4 Argon
56.9
Gemini 4 Carbon
76.8
GPT-6 Astra
74.8
Ebr Bench
Gemini 4 Carbon leads by +11.4
Gemini 4 Argon
64.8
Gemini 4 Carbon
87.6
GPT-6 Astra
76.2
FrontierMath-Tier-4-v2-Private
GPT-6 Astra leads by +2.5
Gemini 4 Argon
78.0
Gemini 4 Carbon
95.1
GPT-6 Astra
97.6
FrontierMath-Tiers-1-3-v2-Private
GPT-6 Astra leads by +3.3
Gemini 4 Argon
90.5
Gemini 4 Carbon
95.3
GPT-6 Astra
98.6
Frontierswe
GPT-6 Astra leads by +4.3
Gemini 4 Argon
55.0
Gemini 4 Carbon
61.2
GPT-6 Astra
65.5
Furniture Assembly
GPT-6 Astra leads by +3.6
Gemini 4 Argon
60.7
Gemini 4 Carbon
67.9
GPT-6 Astra
71.4
GPQA diamond
Gemini 4 Carbon leads by +0.2
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Gemini 4 Argon
99.2
Gemini 4 Carbon
99.3
GPT-6 Astra
98.3
Mystery Game Puzzles
GPT-6 Astra leads by +12.1
Gemini 4 Argon
60.3
Gemini 4 Carbon
70.3
GPT-6 Astra
82.4
OTIS Mock AIME 2024-2025
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 4 Argon
100.0
Gemini 4 Carbon
100.0
GPT-6 Astra
100.0
SimpleBench
Gemini 4 Carbon leads by +1.1
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Gemini 4 Argon
90.0
Gemini 4 Carbon
91.1
GPT-6 Astra
83.8
SimpleQA Verified
Gemini 4 Carbon leads by +1.7
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 4 Argon
76.0
Gemini 4 Carbon
77.7
GPT-6 Astra
75.6
Artificial Analysis · CritPt
GPT-6 Astra leads by +4.6
Gemini 4 Argon
27.1
GPT-6 Astra
31.7
Artificial Analysis · GDPval
Gemini 4 Argon leads by +2.6
Gemini 4 Argon
56.2
GPT-6 Astra
53.6
Artificial Analysis · Humanity's Last Exam
Gemini 4 Argon leads by +2.4
Gemini 4 Argon
57.1
GPT-6 Astra
54.7
Artificial Analysis · Long Context Reasoning
GPT-6 Astra leads by +1.0
Gemini 4 Argon
79.7
GPT-6 Astra
80.7
Artificial Analysis · Quality Index
GPT-6 Astra leads by +0.1
Gemini 4 Argon
52.6
GPT-6 Astra
52.7
Artificial Analysis · SciCode
Gemini 4 Argon leads by +5.3
Gemini 4 Argon
61.8
GPT-6 Astra
56.5
Dtbench
Gemini 4 Argon leads by +0.5
Gemini 4 Argon
96.0
GPT-6 Astra
95.5
Frontiercode
GPT-6 Astra leads by +5.7
Gemini 4 Argon
47.5
GPT-6 Astra
53.3
Lmca
GPT-6 Astra leads by +0.8
Gemini 4 Argon
74.9
GPT-6 Astra
75.7
Mirrorcode
GPT-6 Astra leads by +1.1
Gemini 4 Argon
58.9
GPT-6 Astra
60.0
PostTrainBench
GPT-6 Astra leads by +1.4
Gemini 4 Argon
40.4
GPT-6 Astra
41.9
Proofbench
Gemini 4 Argon
99.0
GPT-6 Astra
99.0
Remote Labor Index
GPT-6 Astra leads by +3.8
Gemini 4 Argon
17.1
GPT-6 Astra
20.8
WeirdML v3
GPT-6 Astra leads by +14.1
Gemini 4 Argon
46.0
GPT-6 Astra
60.0
Full benchmark table
| Benchmark | Gemini 4 Argon | Gemini 4 Carbon | GPT-6 Astra |
|---|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 82.4 | 84.2 | 64.7 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 98.5 | 98.5 | 98.5 |
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. | 91.3 | 94.2 | 95.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 56.9 | 76.8 | 74.8 |
Ebr Bench | 64.8 | 87.6 | 76.2 |
FrontierMath-Tier-4-v2-Private | 78.0 | 95.1 | 97.6 |
FrontierMath-Tiers-1-3-v2-Private | 90.5 | 95.3 | 98.6 |
Frontierswe | 55.0 | 61.2 | 65.5 |
Furniture Assembly | 60.7 | 67.9 | 71.4 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 99.2 | 99.3 | 98.3 |
Mystery Game Puzzles | 60.3 | 70.3 | 82.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 100.0 | 100.0 | 100.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 90.0 | 91.1 | 83.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 76.0 | 77.7 | 75.6 |
Artificial Analysis · CritPt | 27.1 | n/a | 31.7 |
Artificial Analysis · GDPval | 56.2 | n/a | 53.6 |
Artificial Analysis · Humanity's Last Exam | 57.1 | n/a | 54.7 |
Artificial Analysis · Long Context Reasoning | 79.7 | n/a | 80.7 |
Artificial Analysis · Quality Index | 52.6 | n/a | 52.7 |
Artificial Analysis · SciCode | 61.8 | n/a | 56.5 |
Dtbench | 96.0 | n/a | 95.5 |
Frontiercode | 47.5 | n/a | 53.3 |
Lmca | 74.9 | n/a | 75.7 |
Mirrorcode | 58.9 | n/a | 60.0 |
PostTrainBench | 40.4 | n/a | 41.9 |
Proofbench | 99.0 | n/a | 99.0 |
Remote Labor Index | 17.1 | n/a | 20.8 |
WeirdML v3 | 46.0 | n/a | 60.0 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $10.00 | 1.0M tokens (~524 books) | $40.00 | |
| $2.00 | $10.00 | 1.0M tokens (~524 books) | $40.00 | |
| $10.00 | $50.00 | 1.1M tokens (~525 books) | $200.00 |
People also compared