GPT-5.1-Codex vs Kimi K2.5 Thinking

Change models
Kimi K2.5 ThinkingMoonshot AI
IndexAll categories, out of 100.
52.4 ±5.7 best
52.2 ±7.4
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
85% best
75%
SpeedOutput tokens per second.
83/s
—
EloLMArena rating and rank.
—
1445
ReleasedMonth of release.
2025-11
—
Input / 1MUS dollars per 1M input tokens.
$1.25
—
Output / 1MUS dollars per 1M output tokens.
$10.00
—
1K chats2K tokens in, 500 out, each.
$7.50
—
1K documents20K tokens in, 1K out, each.
$35.00
—
1K agent tasks200K tokens in, 20K out, each.
$450.00
—
ContextMaximum tokens in one request.
400K
—
AgenticMulti-step tasks with tools.
55.6 best
47.9
CodingCode writing and repair.
49.3
51.2 best
ReasoningLogic problems and puzzles.
50.7
—
MultimodalTasks with images and text.
49.1
64.1 best
KnowledgeFacts and expert knowledge.
55.4 best
53.2
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
53.7
—
MathMath problems.
—
56.6

Benchmarks

3 shared
Only GPT-5.1-Codex: 12Only Kimi K2.5 Thinking: 19
GPT-5.1-Codex
Kimi K2.5 Thinking
85.5% best30 / 142
83.9%44 / 142
13.1%109 / 130
17.5% best103 / 130
IOI v1Coding
3.7%51 / 61
17.7% best26 / 61