Comparing 2 models

Claude Fable 5.2 vs Gemini 4 Carbon

Benchmarks through

Verdict

Claude Fable 5.2 places higher on 11 of the 11 tasks where both are ranked. Gemini 4 Carbon costs 4.0x less per input token. Gemini 4 Carbon accepts 1.0x more context.

  • Choose Claude Fable 5.2 for Coding (#1), Tool & Function Calling (#1), AI Agents (#1).

Built from the published task rankings below; ranks come from each task's own weighted benchmarks, so a win on one task says nothing about another.

metric
Model 1Claude Fable 5.2
Model 2Gemini 4 Carbon
Best for
Agentic Coding (#1)
a General Assistant (#1)
Provider
anthropic
google
Context
1M
1.0MBest
Input / 1M
$8
$2Best
Output / 1M
$40
$10Best
Tool use
yesBest
yesBest
Vision
yesBest
yesBest
Reasoning
yesBest
yesBest
Open-weight
—
—
Released
Nov 9, 2026
Nov 10, 2026

Measured head to head

Benchmarks both models were run on

EvaluationClaude Fable 5.2Gemini 4 CarbonSource
Design Arena Coding1375 Elo1348 Elosource · 2026-11-15
Design Arena UI1382 Elo1347 Elosource · 2026-11-15
Design Arena Website1352 Elo1319 Elosource · 2026-11-15
FrontierMath Tiers 1-3 (Epoch AI)95.1% ± 1.390.5% ± 1.7source · 2026-11-06
GPQA Diamond (Epoch AI)93.9% ± 1.696.5% ± 1.2source · 2026-11-06
LMArena Agent30.417source · 2026-11-11
LMArena Agent Steerability3319.1source · 2026-11-11
LMArena Agent Task Outcome2218.4source · 2026-11-11
LMArena Agent Tool Hallucination10.5source · 2026-11-11
LMArena WebDev1838 Elo1768 Elosource · 2026-11-11
OTIS Mock AIME (Epoch AI)99.7% ± 0.3100% ± 0source · 2026-11-06
SimpleQA Verified (Epoch AI)94% ± 0.877.7% ± 1.3source · 2026-11-06
Terminal-Bench (4.1)65.8%64.5%source · 2026-11-15

Values are as published by each evaluator, in its own units; rows from different evaluators are not on a common scale.