Inference
EN

Claude Fable 5 vs Gemini 4 Argon

Change models
Claude Fable 5Anthropic
Gemini 4 ArgonGoogle
IndexAll categories, out of 100.
74.0 ±2.7
78.5 ±4.3 best
GapIndex points below the top.
−4.5 · no clear gap
Top
CoverageIndex weight that has results.
95% best
80%
SpeedOutput tokens per second.
56/s best
28/s
EloLMArena rating and rank.
1492
1530 best
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
77.3 best
CodingCode writing and repair.
74.7
76.7 best
ReasoningLogic problems and puzzles.
73.6
79.8 best
MultimodalTasks with images and text.
70.1
—
KnowledgeFacts and expert knowledge.
75.8
76.5 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
64.1
—
MathMath problems.
76.2 best
75.9

Benchmarks

50 shared
Only Claude Fable 5: 42Only Gemini 4 Argon: 36
Claude Fable 5
Gemini 4 Argon
Artificial Analysis Humanity's Last ExamKnowledge
55.5%5 / 222
57.1% best4 / 222
Artificial Analysis Long Context ReasoningReasoning
82.3% best27 / 217
79.7%45 / 217
GPQA diamond · Epoch AIKnowledge
85.9%69 / 217
96.4% best2 / 217
Critical Physics TasksReasoning
28.6% best14 / 214
27.1%17 / 214
OTIS Mock AIME 2024-2025Math
99.7%14 / 200
100.0% best1 / 200
Artificial Analysis Omniscience AccuracyKnowledge
65.4% best4 / 164
48.3%26 / 164
LegalBenchLegal
88.6% best1 / 149
88.3%3 / 149
GDPval-AA normalizedAgentic
55.5%18 / 133
56.2% best14 / 133
ARC-AGI-2 (semi-private)Reasoning
89.2%11 / 131
91.3% best7 / 131
ARC-AGI-1 (semi-private)Reasoning
98.5%1 / 126
98.5%1 / 126
Vibe Code Bench v1.1Coding
90.4%4 / 123
91.9% best3 / 123
ARC-AGI-2 (public eval)Reasoning
96.4%9 / 122
97.1% best6 / 122
ARC-AGI-1 (public eval)Reasoning
97.6%17 / 116
99.5% best1 / 116
Artificial Analysis SciCodeCoding
61.0%6 / 114
61.8% best5 / 114
MedScribeHealthcare
88.5% best12 / 109
87.4%17 / 109
MedCodeHealthcare
56.1%4 / 106
58.8% best3 / 106
SimpleQA VerifiedKnowledge
70.7%9 / 91
76.0% best3 / 91
Chess PuzzlesReasoning
47.0%24 / 90
59.0% best10 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
87.0% best12 / 90
86.0%13 / 90
Legal Research BenchLegal
49.5%9 / 90
54.8% best5 / 90