Claude Fable 5.2 vs Gemini 4 Carbon

Change models
Claude Fable 5.2AnthropicSmartest
IndexAll categories, out of 100.
85.6 ±3.7 best
82.7 ±5.1
GapIndex points below the top.
Top
−2.9 · no clear gap
CoverageIndex weight that has results.
90% best
80%
SpeedOutput tokens per second.
64/s best
31/s
EloLMArena rating and rank.
—
1538
ReleasedMonth of release.
2026-11
2026-10
Input / 1MUS dollars per 1M input tokens.
$8.00
$2.00 best
Output / 1MUS dollars per 1M output tokens.
$40.00
$10.00 best
Batch output / 1MBatch price per 1M output tokens.
$20.00
$5.00 best
1K chats2K tokens in, 500 out, each.
$36.00
$9.00 best
1K documents20K tokens in, 1K out, each.
$200.00
$50.00 best
1K agent tasks200K tokens in, 20K out, each.
$2,400.00
$600.00 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
82.2
83.2 best
CodingCode writing and repair.
86.2 best
80.5
ReasoningLogic problems and puzzles.
86.5 best
83.9
MultimodalTasks with images and text.
82.0
—
KnowledgeFacts and expert knowledge.
87.1 best
79.5
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
79.3 best
77.2

Benchmarks

34 shared
Only Claude Fable 5.2: 37Only Gemini 4 Carbon: 5
Claude Fable 5.2
Gemini 4 Carbon
83.3% best13 / 217
79.3%52 / 217
93.9%14 / 217
96.5% best1 / 217
99.7%14 / 200
100.0% best1 / 200
79.1% best1 / 164
58.3%9 / 164
92.5% best3 / 131
92.1%5 / 131
97.5%5 / 126
97.5%5 / 126
93.8% best1 / 123
89.1%9 / 123
98.8% best1 / 122
97.9%3 / 122
99.0% best2 / 116
98.5%6 / 116
94.0% best1 / 91
77.7%2 / 91
Chess PuzzlesReasoning
79.0% best1 / 90
78.0%2 / 90
95.1% best1 / 90
90.5%6 / 90
46.2%19 / 90
55.8% best1 / 90
32.1%23 / 82
51.8% best1 / 82
90.0% best1 / 80
73.0%5 / 80
100.0% best1 / 72
95.1%4 / 72
65.2% best1 / 61
63.1%4 / 61
33.0 best1 / 54
19.12 / 54
22.0 best1 / 54
18.44 / 54
90.0% best1 / 37
77.5%6 / 37