GPT-5 (high) vs GPT-5 (medium)
Change modelsGPT-5 (high)OpenAI
GPT-5 (medium)OpenAI
IndexAll categories, out of 100.
53.6 ±5.0 best
53.0 ±9.2
GapIndex points below the top.
Top
−0.6 · no clear gap
CoverageIndex weight that has results.
95% best
65%
SpeedOutput tokens per second.
76/s
76/s
EloLMArena rating and rank.
1406
—
ReleasedMonth of release.
2025-08
2025-08
Input / 1MUS dollars per 1M input tokens.
$1.25
$1.25
Output / 1MUS dollars per 1M output tokens.
$10.00
$10.00
1K chats2K tokens in, 500 out, each.
$7.50
$7.50
1K documents20K tokens in, 1K out, each.
$35.00
$35.00
1K agent tasks200K tokens in, 20K out, each.
$450.00
$450.00
ContextMaximum tokens in one request.
400K
400K
AgenticMulti-step tasks with tools.
49.7
60.2 best
CodingCode writing and repair.
59.5
—
ReasoningLogic problems and puzzles.
54.4 best
48.1
MultimodalTasks with images and text.
51.7
51.8 best
KnowledgeFacts and expert knowledge.
56.1 best
54.9
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
57.3 best
55.0
MathMath problems.
49.2
—
Benchmarks
8 sharedOnly GPT-5 (high): 7
GPT-5 (high)
GPT-5 (medium)