GPT-5 (medium) vs Kimi K2.5 (Reasoning)
Change modelsGPT-5 (medium)OpenAI
Kimi K2.5 (Reasoning)Moonshot AI
IndexAll categories, out of 100.
53.0 ±9.2
53.0 ±4.0
GapIndex points below the top.
Top
Top
CoverageIndex weight that has results.
65%
95% best
SpeedOutput tokens per second.
76/s
—
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2025-08
2026-01
Input / 1MUS dollars per 1M input tokens.
$1.25
$0.60 best
Output / 1MUS dollars per 1M output tokens.
$10.00
$3.00 best
1K chats2K tokens in, 500 out, each.
$7.50
$2.70 best
1K documents20K tokens in, 1K out, each.
$35.00
$15.00 best
1K agent tasks200K tokens in, 20K out, each.
$450.00
$180.00 best
ContextMaximum tokens in one request.
400K best
128K
AgenticMulti-step tasks with tools.
60.2 best
50.6
CodingCode writing and repair.
—
56.0
ReasoningLogic problems and puzzles.
48.1
52.1 best
MultimodalTasks with images and text.
51.8
53.9 best
KnowledgeFacts and expert knowledge.
54.9
55.5 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
55.0 best
54.6
MathMath problems.
—
52.2
Benchmarks
8 sharedOnly Kimi K2.5 (Reasoning): 15
GPT-5 (medium)
Kimi K2.5 (Reasoning)