Compare · ModelsLive · 3 picked · head to head
Claude Fable 5.2 vs Gemini 4 Carbon vs GPT-6.1 Terra
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Fable 5.2 wins on 10/15 benchmarks
Claude Fable 5.2 wins 10 of 15 shared benchmarks. Leads in general · math · knowledge.
Category leads
agentic·Gemini 4 Carbongeneral·Claude Fable 5.2math·Claude Fable 5.2knowledge·Claude Fable 5.2reasoning·Claude Fable 5.2coding·Claude Fable 5.2
Hype vs Reality
Attention vs performance
Claude Fable 5.2
#2 by perf·#5 by attention
Gemini 4 Carbon
#1 by perf·#8 by attention
GPT-6.1 Terra
#14 by perf·#2 by attention
Best value
GPT-6.1 Terra
1.7x better value than Gemini 4 Carbon
Claude Fable 5.2
3.5 pts/$
$24.00/M
Gemini 4 Carbon
14.3 pts/$
$6.00/M
GPT-6.1 Terra
24.1 pts/$
$3.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
Google DeepMind
$3.61T·Tier 1
OpenAI
$840.0B·Tier 1
Head to head
15 benchmarks · 3 models
Claude Fable 5.2Gemini 4 CarbonGPT-6.1 Terra
APEX-Agents
Gemini 4 Carbon leads by +4.1
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Fable 5.2
80.1
Gemini 4 Carbon
84.2
GPT-6.1 Terra
68.3
Ebr Bench
Claude Fable 5.2 leads by +2.9
Claude Fable 5.2
90.5
Gemini 4 Carbon
87.6
GPT-6.1 Terra
78.1
FrontierMath-Tier-4-v2-Private
Claude Fable 5.2 leads by +4.9
Claude Fable 5.2
100.0
Gemini 4 Carbon
95.1
GPT-6.1 Terra
92.7
FrontierMath-Tiers-1-3-v2-Private
Claude Fable 5.2 leads by +4.3
Claude Fable 5.2
100.0
Gemini 4 Carbon
95.3
GPT-6.1 Terra
95.7
Furniture Assembly
Claude Fable 5.2 leads by +15.5
Claude Fable 5.2
85.7
Gemini 4 Carbon
67.9
GPT-6.1 Terra
70.2
GPQA diamond
Gemini 4 Carbon leads by +3.1
Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs.
Claude Fable 5.2
95.7
Gemini 4 Carbon
99.3
GPT-6.1 Terra
96.3
Mystery Game Puzzles
Claude Fable 5.2 leads by +17.6
Claude Fable 5.2
89.0
Gemini 4 Carbon
70.3
GPT-6.1 Terra
71.4
OTIS Mock AIME 2024-2025
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Fable 5.2
99.7
Gemini 4 Carbon
100.0
GPT-6.1 Terra
100.0
SimpleBench
Claude Fable 5.2 leads by +2.8
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Fable 5.2
93.9
Gemini 4 Carbon
91.1
GPT-6.1 Terra
66.6
SimpleQA Verified
Claude Fable 5.2 leads by +16.3
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Fable 5.2
94.0
Gemini 4 Carbon
77.7
GPT-6.1 Terra
57.3
ARC-AGI
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Fable 5.2
98.5
Gemini 4 Carbon
98.5
ARC-AGI-2
Gemini 4 Carbon leads by +0.4
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Claude Fable 5.2
93.8
Gemini 4 Carbon
94.2
Chess Puzzles
Claude Fable 5.2 leads by +1.1
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Fable 5.2
77.9
Gemini 4 Carbon
76.8
Frontierswe
Claude Fable 5.2 leads by +8.7
Claude Fable 5.2
69.9
Gemini 4 Carbon
61.2
SpireClimb
GPT-6.1 Terra leads by +2.7
Claude Fable 5.2
19.6
GPT-6.1 Terra
22.3
Full benchmark table
| Benchmark | Claude Fable 5.2 | Gemini 4 Carbon | GPT-6.1 Terra |
|---|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 80.1 | 84.2 | 68.3 |
Ebr Bench | 90.5 | 87.6 | 78.1 |
FrontierMath-Tier-4-v2-Private | 100.0 | 95.1 | 92.7 |
FrontierMath-Tiers-1-3-v2-Private | 100.0 | 95.3 | 95.7 |
Furniture Assembly | 85.7 | 67.9 | 70.2 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 95.7 | 99.3 | 96.3 |
Mystery Game Puzzles | 89.0 | 70.3 | 71.4 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 99.7 | 100.0 | 100.0 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 93.9 | 91.1 | 66.6 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 94.0 | 77.7 | 57.3 |
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 | n/a |
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. | 93.8 | 94.2 | n/a |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 77.9 | 76.8 | n/a |
Frontierswe | 69.9 | 61.2 | n/a |
SpireClimb | 19.6 | n/a | 22.3 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $8.00 | $40.00 | 1.0M tokens (~500 books) | $160.00 | |
| $2.00 | $10.00 | 1.0M tokens (~524 books) | $40.00 | |
| $1.00 | $5.00 | 1.1M tokens (~525 books) | $20.00 |
People also compared