Compare · ModelsLive · 2 picked · head to head
Gemini 3.1 Pro Preview vs Claude Fable 5.1
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Fable 5.1 wins on 20/24 benchmarks
Claude Fable 5.1 wins 20 of 24 shared benchmarks. Leads in speed · agentic · general.
Category leads
speed·Claude Fable 5.1agentic·Claude Fable 5.1reasoning·Gemini 3.1 Pro Previewknowledge·Gemini 3.1 Pro Previewgeneral·Claude Fable 5.1math·Claude Fable 5.1coding·Claude Fable 5.1
Hype vs Reality
Attention vs performance
Gemini 3.1 Pro Preview
#137 by perf·#8 by attention
Claude Fable 5.1
#16 by perf·#12 by attention
Best value
Gemini 3.1 Pro Preview
2.9x better value than Claude Fable 5.1
Gemini 3.1 Pro Preview
6.9 pts/$
$7.00/M
Claude Fable 5.1
2.4 pts/$
$30.00/M
Vendor risk
Who is behind the model
Google DeepMind
$4.20T·Tier 1
Anthropic
$965.0B·Tier 1
Head to head
24 benchmarks · 2 models
Gemini 3.1 Pro PreviewClaude Fable 5.1
Artificial Analysis · CritPt
Claude Fable 5.1 leads by +12.0
Gemini 3.1 Pro Preview
17.7
Claude Fable 5.1
29.7
Artificial Analysis · GDPval
Claude Fable 5.1 leads by +47.9
Gemini 3.1 Pro Preview
13.8
Claude Fable 5.1
61.7
Artificial Analysis · GPQA Diamond
Gemini 3.1 Pro Preview leads by +0.4
Gemini 3.1 Pro Preview
94.1
Claude Fable 5.1
93.7
Artificial Analysis · Humanity's Last Exam
Claude Fable 5.1 leads by +12.1
Gemini 3.1 Pro Preview
47.0
Claude Fable 5.1
59.1
Artificial Analysis · Long Context Reasoning
Claude Fable 5.1 leads by +3.3
Gemini 3.1 Pro Preview
82.0
Claude Fable 5.1
85.3
Artificial Analysis · Quality Index
Claude Fable 5.1 leads by +23.6
Gemini 3.1 Pro Preview
29.7
Claude Fable 5.1
53.4
Artificial Analysis · SciCode
Claude Fable 5.1 leads by +4.4
Gemini 3.1 Pro Preview
58.7
Claude Fable 5.1
63.1
APEX-Agents
Claude Fable 5.1 leads by +33.3
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Gemini 3.1 Pro Preview
35.3
Claude Fable 5.1
68.6
ARC-AGI
Gemini 3.1 Pro Preview leads by +0.5
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Gemini 3.1 Pro Preview
98.0
Claude Fable 5.1
97.5
ARC-AGI-2
Claude Fable 5.1 leads by +12.9
ARC-AGI-2 · the second iteration of the Abstraction and Reasoning Corpus, testing novel pattern recognition and abstract reasoning without prior training data.
Gemini 3.1 Pro Preview
77.1
Claude Fable 5.1
90.0
Chess Puzzles
Gemini 3.1 Pro Preview leads by +8.4
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Gemini 3.1 Pro Preview
52.6
Claude Fable 5.1
44.2
Dtbench
Claude Fable 5.1 leads by +0.9
Gemini 3.1 Pro Preview
95.1
Claude Fable 5.1
96.0
Ebr Bench
Claude Fable 5.1 leads by +42.9
Gemini 3.1 Pro Preview
14.3
Claude Fable 5.1
57.1
FrontierMath-Tier-4-v2-Private
Claude Fable 5.1 leads by +61.0
Gemini 3.1 Pro Preview
26.8
Claude Fable 5.1
87.8
FrontierMath-Tiers-1-3-v2-Private
Claude Fable 5.1 leads by +30.5
Gemini 3.1 Pro Preview
59.6
Claude Fable 5.1
90.2
Furniture Assembly
Claude Fable 5.1 leads by +57.1
Gemini 3.1 Pro Preview
0.0
Claude Fable 5.1
57.1
HLE
Claude Fable 5.1 leads by +0.1
HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%.
Gemini 3.1 Pro Preview
43.7
Claude Fable 5.1
43.8
Lmca
Claude Fable 5.1 leads by +13.7
Gemini 3.1 Pro Preview
63.3
Claude Fable 5.1
77.0
Mirrorcode
Claude Fable 5.1 leads by +64.4
Gemini 3.1 Pro Preview
8.9
Claude Fable 5.1
73.3
Mystery Game Puzzles
Claude Fable 5.1 leads by +26.4
Gemini 3.1 Pro Preview
27.3
Claude Fable 5.1
53.7
OTIS Mock AIME 2024-2025
Claude Fable 5.1 leads by +4.4
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Gemini 3.1 Pro Preview
95.6
Claude Fable 5.1
100.0
Proofbench
Claude Fable 5.1 leads by +74.0
Gemini 3.1 Pro Preview
26.0
Claude Fable 5.1
100.0
SimpleQA Verified
Gemini 3.1 Pro Preview leads by +2.7
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Gemini 3.1 Pro Preview
73.5
Claude Fable 5.1
70.8
WeirdML
Claude Fable 5.1 leads by +20.8
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Gemini 3.1 Pro Preview
72.1
Claude Fable 5.1
92.9
Full benchmark table
| Benchmark | Gemini 3.1 Pro Preview | Claude Fable 5.1 |
|---|---|---|
Artificial Analysis · CritPt | 17.7 | 29.7 |
Artificial Analysis · GDPval | 13.8 | 61.7 |
Artificial Analysis · GPQA Diamond | 94.1 | 93.7 |
Artificial Analysis · Humanity's Last Exam | 47.0 | 59.1 |
Artificial Analysis · Long Context Reasoning | 82.0 | 85.3 |
Artificial Analysis · Quality Index | 29.7 | 53.4 |
Artificial Analysis · SciCode | 58.7 | 63.1 |
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 35.3 | 68.6 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 98.0 | 97.5 |
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. | 77.1 | 90.0 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 52.6 | 44.2 |
Dtbench | 95.1 | 96.0 |
Ebr Bench | 14.3 | 57.1 |
FrontierMath-Tier-4-v2-Private | 26.8 | 87.8 |
FrontierMath-Tiers-1-3-v2-Private | 59.6 | 90.2 |
Furniture Assembly | 0.0 | 57.1 |
HLE HLE (Humanity's Last Exam) · a reasoning benchmark designed to be the hardest public evaluation of AI. Questions span mathematics, physics, philosophy, and logic · curated to be at or beyond the frontier of human expert capability. Tested with and without tool augmentation. Claude Opus 4.7 scores 46.9% without tools and 54.7% with tools · making it one of the few benchmarks where the top score is below 60%. | 43.7 | 43.8 |
Lmca | 63.3 | 77.0 |
Mirrorcode | 8.9 | 73.3 |
Mystery Game Puzzles | 27.3 | 53.7 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 95.6 | 100.0 |
Proofbench | 26.0 | 100.0 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 73.5 | 70.8 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 72.1 | 92.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $2.00 | $12.00 | 1.0M tokens (~524 books) | $45.00 | |
| $10.00 | $50.00 | 1.0M tokens (~500 books) | $200.00 |
People also compared