Qwen3.5 Plus Thinking vs Grok 4.1 Fast (Reasoning)

Change models
IndexAll categories, out of 100.
47.7 ±8.2 best
47.1 ±3.9
GapIndex points below the top.
Top
−0.6 · no clear gap
CoverageIndex weight that has results.
75%
100% best
SpeedOutput tokens per second.
—
—
EloLMArena rating and rank.
—
1408
ReleasedMonth of release.
—
2025-11
Input / 1MUS dollars per 1M input tokens.
—
—
Output / 1MUS dollars per 1M output tokens.
—
—
1K chats2K tokens in, 500 out, each.
—
—
1K documents20K tokens in, 1K out, each.
—
—
1K agent tasks200K tokens in, 20K out, each.
—
—
ContextMaximum tokens in one request.
—
2M
AgenticMulti-step tasks with tools.
51.3 best
48.4
CodingCode writing and repair.
53.3 best
42.9
ReasoningLogic problems and puzzles.
—
49.6
MultimodalTasks with images and text.
5.0
43.3 best
KnowledgeFacts and expert knowledge.
55.4 best
48.2
MultilingualTasks in many languages.
—
59.2
InstructionTasks with strict prompt rules.
—
38.2
MathMath problems.
52.4
54.9 best

Benchmarks

14 shared
Only Grok 4.1 Fast (Reasoning): 27
Qwen3.5 Plus Thinking
Grok 4.1 Fast (Reasoning)
85.1% best25 / 149
82.5%69 / 149
85.3% best33 / 142
80.6%66 / 142
87.4% best38 / 135
84.3%51 / 135
MMLU ProKnowledge
87.2% best32 / 135
84.2%64 / 135
CorpFin v2Finance
65.3%28 / 131
66.0% best23 / 131
15.7% best104 / 130
1.2%125 / 130
60.8% best64 / 97
42.6%89 / 97
AIMEMath
86.0%33 / 95
91.9% best21 / 95
MedQAHealthcare
95.2% best13 / 95
92.1%35 / 95
MMMU ProMultimodal
22.8%92 / 92
72.7% best62 / 92
SWE-benchCoding
71.2% best59 / 87
41.4%84 / 87
41.6% best29 / 66
24.7%52 / 66
59.7%25 / 53
60.5% best24 / 53
54.5% best16 / 51
52.4%23 / 51