Skip to content
Leaderboard / OpenAI

GPT-5.4 nano

OpenAI High CLOSED Proprietary
Verdict

GPT-5.4 nano is behind the leaders: #189 of 406 by SIscope score. Strongest at math, weaker at reasoning & knowledge. Cheaper than 70% of priced models.

SIscope score
50
#189 of 406
Human-vote rank
#185
of 419 · score 1372
Price per 1M tokens, blended
$0.46
$0.2 in · $1.25 out
Context window
400K
400,000 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsno data yetReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding59
Agents & toolsNo data yet–
Reasoning & knowledge39
Math61
Chat quality56
Following instructions57

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

136513731380Oct 20: 1372 (rank #183)Oct 22: 1372.2 (rank #184)Oct 28: 1372 (rank #185)Oct 20Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond78%
PhD-level science multiple choice questions
Better than 55% of 333 tested models. Run Apr 14, 2026.
SimpleQA Verified12%
Short factual questions; rewards not making things up
Better than 3% of 100 tested models. Run Aug 27, 2026.
OTIS Mock AIME 2024-2588%
Competition maths problems in the style of AIME
Better than 70% of 313 tested models. Run Apr 14, 2026.
FrontierMath Tiers 1-345%
Very hard, unpublished research-style maths problems
Better than 50% of 122 tested models. Run Jun 12, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#185 of 41913721368–137662,314Oct 28, 2026
WebDev arena
Building web apps, blind human votes
Not on this board yet
Vision arena
Questions about images, blind human votes
#96 of 16711961189–120225,624Oct 19, 2026
Agent arena
Real multi-step agent sessions
Not on this board yet

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$0.2
per 1M tokens
Output
$1.25
per 1M tokens
Blended
$0.46
3 in : 1 out, per 1M
Context window
400K
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (openai/gpt-5.4-nano), Nov 4, 2026. Providers can charge differently.