Grok 4 Fast (Reasoning) vs Qwen3 Max
Change models
Qwen3 MaxAlibaba
IndexAll categories, out of 100.
45.3 ±4.3 best
45.0 ±5.1
GapIndex points below the top.
Top
−0.3 · no clear gap
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
—
62/s
EloLMArena rating and rank.
1400
1413 best
ReleasedMonth of release.
2025-09
2025-09
Input / 1MUS dollars per 1M input tokens.
—
$0.78
Output / 1MUS dollars per 1M output tokens.
—
$3.90
1K chats2K tokens in, 500 out, each.
—
$3.51
1K documents20K tokens in, 1K out, each.
—
$19.50
1K agent tasks200K tokens in, 20K out, each.
—
$234.00
ContextMaximum tokens in one request.
2M best
262K
AgenticMulti-step tasks with tools.
42.9
47.1 best
CodingCode writing and repair.
44.9
46.3 best
ReasoningLogic problems and puzzles.
47.8 best
40.3
MultimodalTasks with images and text.
42.7
—
KnowledgeFacts and expert knowledge.
46.6
48.1 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
36.2 best
30.2
MathMath problems.
55.4 best
46.3
Benchmarks
24 sharedOnly Grok 4 Fast (Reasoning): 11Only Qwen3 Max: 10
Grok 4 Fast (Reasoning)
Qwen3 Max