Inference
EN

Claude Fable 5 vs GPT-6 Sol

Change models
Claude Fable 5Anthropic
GPT-6 SolOpenAI
IndexAll categories, out of 100.
74.0 ±2.7 best
72.8 ±2.2
GapIndex points below the top.
Top
−1.2 · no clear gap
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
56/s
61/s best
EloLMArena rating and rank.
1492 best
1397
ReleasedMonth of release.
2026-06
2026-09
Input / 1MUS dollars per 1M input tokens.
$10.00
$2.00 best
Output / 1MUS dollars per 1M output tokens.
$50.00
$10.00 best
Batch output / 1MBatch price per 1M output tokens.
$25.00
$5.00 best
1K chats2K tokens in, 500 out, each.
$45.00
$9.00 best
1K documents20K tokens in, 1K out, each.
$250.00
$50.00 best
1K agent tasks200K tokens in, 20K out, each.
$3,000.00
$600.00 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
71.4 best
68.6
CodingCode writing and repair.
74.7 best
73.9
ReasoningLogic problems and puzzles.
73.6
76.1 best
MultimodalTasks with images and text.
70.1
70.1
KnowledgeFacts and expert knowledge.
75.8 best
71.9
MultilingualTasks in many languages.
—
65.5
InstructionTasks with strict prompt rules.
64.1
65.3 best
MathMath problems.
76.2 best
75.8

Benchmarks

54 shared
Only Claude Fable 5: 38Only GPT-6 Sol: 40
Claude Fable 5
GPT-6 Sol
Artificial Analysis Humanity's Last ExamKnowledge
55.5% best5 / 222
47.9%18 / 222
Artificial Analysis Long Context ReasoningReasoning
82.3%27 / 217
83.7% best10 / 217
GPQA diamond · Epoch AIKnowledge
85.9%69 / 217
94.3% best10 / 217
Critical Physics TasksReasoning
28.6%14 / 214
30.9% best8 / 214
OTIS Mock AIME 2024-2025Math
99.7%14 / 200
100.0% best1 / 200
Artificial Analysis Omniscience AccuracyKnowledge
65.4% best4 / 164
54.5%13 / 164
GDPval-AA normalizedAgentic
55.5% best18 / 133
50.4%27 / 133
ARC-AGI-2 (semi-private)Reasoning
89.2%11 / 131
89.6% best10 / 131
ARC-AGI-1 (semi-private)Reasoning
98.5% best1 / 126
95.5%17 / 126
Vibe Code Bench v1.1Coding
90.4% best4 / 123
87.8%12 / 123
ARC-AGI-2 (public eval)Reasoning
96.4% best9 / 122
90.0%17 / 122
ARC-AGI-1 (public eval)Reasoning
97.6%17 / 116
97.8% best14 / 116
Artificial Analysis SciCodeCoding
61.0% best6 / 114
57.6%18 / 114
MedScribeHealthcare
88.5% best12 / 109
82.0%52 / 109
MedCodeHealthcare
56.1% best4 / 106
47.1%36 / 106
SimpleQA VerifiedKnowledge
70.7% best9 / 91
60.7%17 / 91
Chess PuzzlesReasoning
47.0%24 / 90
56.0% best15 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
87.0%12 / 90
89.8% best8 / 90
Legal Research BenchLegal
49.5% best9 / 90
28.8%57 / 90
Tax Agent BenchFinance
38.3% best12 / 82
15.0%62 / 82