Claude Fable 5.2 vs Qwen4 Max

Change models
Claude Fable 5.2AnthropicSmartest
Qwen4 MaxAlibaba
IndexAll categories, out of 100.
85.6 ±3.5 best
72.0 ±8.0
GapIndex points below the top.
Top
−13.6
CoverageIndex weight that has results.
90% best
65%
SpeedOutput tokens per second.
80/s best
40/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2026-11
2026-11
Input / 1MUS dollars per 1M input tokens.
$8.00
$1.50 best
Output / 1MUS dollars per 1M output tokens.
$40.00
$4.50 best
Batch output / 1MBatch price per 1M output tokens.
$20.00
—
1K chats2K tokens in, 500 out, each.
$36.00
$5.25 best
1K documents20K tokens in, 1K out, each.
$200.00
$34.50 best
1K agent tasks200K tokens in, 20K out, each.
$2,400.00
$390.00 best
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
82.8 best
75.2
CodingCode writing and repair.
85.6 best
67.0
ReasoningLogic problems and puzzles.
86.6
—
MultimodalTasks with images and text.
83.2 best
72.1
KnowledgeFacts and expert knowledge.
87.2 best
70.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
78.9
—

Benchmarks

15 shared
Only Claude Fable 5.2: 60Only Qwen4 Max: 4
Claude Fable 5.2
Qwen4 Max
79.1% best1 / 167
43.7%42 / 167
93.8% best1 / 125
86.7%14 / 125
46.2% best20 / 92
41.3%31 / 92
32.1%25 / 84
40.4% best12 / 84
74.2% best1 / 79
51.7%22 / 79
65.2% best1 / 63
48.5%14 / 63
DeepSWEAgentic
74.3%10 / 61
75.7% best5 / 61
25.5 best1 / 56
8.415 / 56
73.2% best1 / 54
51.7%8 / 54
5.0%29 / 37
21.3% best7 / 37
76.4% best3 / 24
72.6%6 / 24
24.2%14 / 23
26.2% best9 / 23
59.6% best4 / 23
48.5%13 / 23
85.6% best4 / 21
31.5%12 / 21
73.4% best1 / 11
54.2%7 / 11