GPT-5 vs Qwen3.5 397B

Change models
GPT-5OpenAI
IndexAll categories, out of 100.
53.5 ±3.4 best
53.1 ±2.4
GapIndex points below the top.
Top
−0.4 · no clear gap
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
77/s best
53/s
EloLMArena rating and rank.
1406
—
ReleasedMonth of release.
2025-08
2026-02
Input / 1MUS dollars per 1M input tokens.
$1.25
$0.60 best
Output / 1MUS dollars per 1M output tokens.
$10.00
$3.60 best
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$7.50
$3.00 best
1K documents20K tokens in, 1K out, each.
$35.00
$15.60 best
1K agent tasks200K tokens in, 20K out, each.
$450.00
$192.00 best
ContextMaximum tokens in one request.
400K best
128K
AgenticMulti-step tasks with tools.
49.8
52.6 best
CodingCode writing and repair.
51.3
52.6 best
ReasoningLogic problems and puzzles.
54.3 best
51.3
MultimodalTasks with images and text.
55.9 best
54.7
KnowledgeFacts and expert knowledge.
57.4 best
53.2
MultilingualTasks in many languages.
—
54.9
InstructionTasks with strict prompt rules.
56.5 best
54.8
MathMath problems.
54.8 best
52.7

Benchmarks

8 shared
Only GPT-5: 52Only Qwen3.5 397B: 62
GPT-5
Qwen3.5 397B
28.5%84 / 216
29.0% best83 / 216
78.2% best68 / 211
77.3%76 / 211
5.7% best77 / 207
1.7%105 / 207
85.4%73 / 193
89.3% best45 / 193
84.8%55 / 143
95.6% best17 / 143
73.1%31 / 140
78.8% best6 / 140
74.2%57 / 109
77.3% best41 / 109
90.7%
90.0%