o1 vs Qwen3.5-4B
Change modelso1OpenAI
Qwen3.5-4BAlibaba
IndexAll categories, out of 100.
44.4 ±4.1 best
44.2 ±7.4
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
95% best
80%
SpeedOutput tokens per second.
97/s best
20/s
EloLMArena rating and rank.
1366
—
ReleasedMonth of release.
2024-12
—
Input / 1MUS dollars per 1M input tokens.
$15.00
—
Output / 1MUS dollars per 1M output tokens.
$60.00
—
1K chats2K tokens in, 500 out, each.
$60.00
—
1K documents20K tokens in, 1K out, each.
$360.00
—
1K agent tasks200K tokens in, 20K out, each.
$4,200.00
—
ContextMaximum tokens in one request.
200K
262K best
AgenticMulti-step tasks with tools.
44.3
63.7 best
CodingCode writing and repair.
31.0
—
ReasoningLogic problems and puzzles.
43.6
43.7 best
MultimodalTasks with images and text.
55.8 best
40.2
KnowledgeFacts and expert knowledge.
46.7 best
41.0
MultilingualTasks in many languages.
—
35.6
InstructionTasks with strict prompt rules.
51.7 best
37.7
MathMath problems.
43.8 best
41.3
Benchmarks
7 sharedOnly o1: 22Only Qwen3.5-4B: 9
o1
Qwen3.5-4B