Compare · ModelsLive · 2 picked · head to head
Claude Fable 5 vs Claude Opus 4.8
Side by side · benchmarks, pricing, and signals you can act on.
Winner summary
Claude Fable 5 wins on 22/24 benchmarks
Claude Fable 5 wins 22 of 24 shared benchmarks. Leads in agentic · reasoning · arena.
Category leads
agentic·Claude Fable 5reasoning·Claude Fable 5arena·Claude Fable 5knowledge·Claude Fable 5coding·Claude Fable 5general·Claude Fable 5math·Claude Fable 5
Hype vs Reality
Attention vs performance
Claude Fable 5
#27 by perf·#12 by attention
Claude Opus 4.8
#75 by perf·#9 by attention
Best value
Claude Opus 4.8
1.6x better value than Claude Fable 5
Claude Fable 5
2.3 pts/$
$30.00/M
Claude Opus 4.8
3.7 pts/$
$15.00/M
Vendor risk
Who is behind the model
Anthropic
$965.0B·Tier 1
Anthropic
$965.0B·Tier 1
Head to head
24 benchmarks · 2 models
Claude Fable 5Claude Opus 4.8
APEX-Agents
Claude Fable 5 leads by +14.7
APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments.
Claude Fable 5
63.6
Claude Opus 4.8
48.9
ARC-AGI
Claude Fable 5 leads by +6.0
ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization.
Claude Fable 5
98.5
Claude Opus 4.8
92.5
ARC-AGI-2
Claude Fable 5 leads by +17.1
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
89.2
Claude Opus 4.8
72.1
Chatbot Arena Elo · Coding
Claude Fable 5 leads by +119.8
Claude Fable 5
1653.9
Claude Opus 4.8
1534.1
Chatbot Arena Elo · Overall
Claude Fable 5 leads by +33.0
Claude Fable 5
1507.6
Claude Opus 4.8
1474.5
Chess Puzzles
Claude Fable 5 leads by +7.4
Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.
Claude Fable 5
37.9
Claude Opus 4.8
30.6
Deepswe
Claude Fable 5 leads by +10.9
Claude Fable 5
69.9
Claude Opus 4.8
59.0
Dtbench
Claude Fable 5 leads by +5.8
Claude Fable 5
97.3
Claude Opus 4.8
91.5
Ebr Bench
Claude Fable 5 leads by +11.0
Claude Fable 5
39.5
Claude Opus 4.8
28.6
Frontiercode
Claude Fable 5 leads by +7.0
Claude Fable 5
53.5
Claude Opus 4.8
46.5
FrontierMath-Tier-4-v2-Private
Claude Fable 5 leads by +34.1
Claude Fable 5
90.2
Claude Opus 4.8
56.1
FrontierMath-Tiers-1-3-v2-Private
Claude Fable 5 leads by +7.0
Claude Fable 5
87.0
Claude Opus 4.8
80.0
Furniture Assembly
Claude Opus 4.8 leads by +9.5
Claude Fable 5
8.3
Claude Opus 4.8
17.9
GPQA diamond
Claude Opus 4.8 leads by +6.9
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
81.1
Claude Opus 4.8
88.0
Lmca
Claude Fable 5 leads by +4.3
Claude Fable 5
71.9
Claude Opus 4.8
67.7
Mystery Game Puzzles
Claude Fable 5 leads by +17.6
Claude Fable 5
47.1
Claude Opus 4.8
29.5
OTIS Mock AIME 2024-2025
Claude Fable 5 leads by +1.7
OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills.
Claude Fable 5
100.0
Claude Opus 4.8
98.3
PostTrainBench
Claude Fable 5 leads by +7.9
Claude Fable 5
41.8
Claude Opus 4.8
33.8
Proofbench
Claude Fable 5 leads by +26.0
Claude Fable 5
95.0
Claude Opus 4.8
69.0
Remote Labor Index
Claude Fable 5 leads by +7.8
Claude Fable 5
16.1
Claude Opus 4.8
8.3
SimpleBench
Claude Fable 5 leads by +20.5
SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking.
Claude Fable 5
78.3
Claude Opus 4.8
57.8
SimpleQA Verified
Claude Fable 5 leads by +17.7
SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information.
Claude Fable 5
70.7
Claude Opus 4.8
53.0
Surface Evolver Bench
Claude Fable 5 leads by +7.5
Claude Fable 5
95.0
Claude Opus 4.8
87.5
WeirdML
Claude Fable 5 leads by +9.0
WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns.
Claude Fable 5
91.9
Claude Opus 4.8
82.9
Full benchmark table
| Benchmark | Claude Fable 5 | Claude Opus 4.8 |
|---|---|---|
APEX-Agents APEX-Agents · evaluates AI agents on complex, multi-step tasks requiring planning, tool use, and autonomous decision-making in realistic environments. | 63.6 | 48.9 |
ARC-AGI ARC-AGI · the original Abstraction and Reasoning Corpus, testing whether AI can solve novel visual pattern recognition tasks without memorization. | 98.5 | 92.5 |
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. | 89.2 | 72.1 |
Chatbot Arena Elo · Coding | 1653.9 | 1534.1 |
Chatbot Arena Elo · Overall | 1507.6 | 1474.5 |
Chess Puzzles Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities. | 37.9 | 30.6 |
Deepswe | 69.9 | 59.0 |
Dtbench | 97.3 | 91.5 |
Ebr Bench | 39.5 | 28.6 |
Frontiercode | 53.5 | 46.5 |
FrontierMath-Tier-4-v2-Private | 90.2 | 56.1 |
FrontierMath-Tiers-1-3-v2-Private | 87.0 | 80.0 |
Furniture Assembly | 8.3 | 17.9 |
GPQA diamond Graduate-Level Google-Proof QA (Diamond set) · expert-crafted questions in physics, biology, and chemistry that are difficult even for domain PhDs. | 81.1 | 88.0 |
Lmca | 71.9 | 67.7 |
Mystery Game Puzzles | 47.1 | 29.5 |
OTIS Mock AIME 2024-2025 OTIS Mock AIME 2024-2025 · simulated American Invitational Mathematics Examination problems testing advanced problem-solving skills. | 100.0 | 98.3 |
PostTrainBench | 41.8 | 33.8 |
Proofbench | 95.0 | 69.0 |
Remote Labor Index | 16.1 | 8.3 |
SimpleBench SimpleBench · tests fundamental reasoning capabilities with straightforward problems designed to expose gaps in basic logical and spatial thinking. | 78.3 | 57.8 |
SimpleQA Verified SimpleQA Verified · short factual questions with verified answers, measuring factual accuracy and the tendency to hallucinate or provide incorrect information. | 70.7 | 53.0 |
Surface Evolver Bench | 95.0 | 87.5 |
WeirdML WeirdML · tests models on unusual and adversarial machine learning tasks that require creative problem-solving beyond standard patterns. | 91.9 | 82.9 |
Pricing · per 1M tokens · projected $/mo at 10M tokens
| Model | Input | Output | Context | Projected $/mo |
|---|---|---|---|---|
| $10.00 | $50.00 | 1.0M tokens (~500 books) | $200.00 | |
| $5.00 | $25.00 | 1.0M tokens (~500 books) | $100.00 |