GPT-5.4 nano (medium)
OpenAIGPTProprietary
Description
GPT-5.4 nano is OpenAI's cheapest GPT-5.4-class model for simple high-volume tasks. It is optimized for classification, data extraction, ranking, and lightweight coding subagents, while preserving strong reasoning, multimodal understanding, and tool use for latency-sensitive workloads.
Release Date
2026-03-17
Parameters
—
Context Length
400K
Modalities
file, image, text
Capability Radar
33
general
36
coding
76
reasoning
51
scienceest.
60
agents
85
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agents & Tools | 81 | 46.0 | LS |
| Code Ranking | 94 | 62.0 | AA |
| General Ranking | 124 | 62.0 | AA |
| Multimodal Ranking | 43 | 76.0 | LS |
| Reasoning | 60 | 62.0 | LS |
| Science | 114 | 58.0 | AA |
Benchmark Scores (LLM Stats)
Agents
MCP Atlas
56.1%SR
SWE-Bench Pro
52.4%SR
Terminal-Bench 2.0
46.3%SR
OSWorld-Verified
39.0%SR
Toolathlon
35.5%SR
Biology
GPQA
82.8%SR
Communication
Tau2 Telecom
92.5%SR
General
MMMU-Pro
66.1%SR
MRCR v2 (8-needle)
33.1%SR
Math
Humanity's Last Exam
24.3%SR
Multimodal
OmniDocBench 1.5
75.8%SR
Reasoning
Graphwalks BFS <128k
73.4%SR
Graphwalks parents <128k
50.8%SR
AA Evaluation Indices
Intelligence Index38.1
Coding Index35.0
Gpqa0.8
Ifbench0.6
Lcr0.6
Tau20.5
Scicode0.4
Terminalbench Hard0.3
Hle0.1
LLM Stats Category Scores
Robotics33
Spatial Reasoning11
Multimodal9
Vision7
Reasoning3
Communication90
Structured Output80
Biology80
Chemistry80
Physics80
Tool Calling60
General60
Agents50
Code50
Long Context40
Healthcare30
Math20
Pricing
Input Price$0.2 / 1M tokens
Output Price$1.25 / 1M tokens
Blended Price (3:1)$0.463 / 1M tokens
Speed
Tokens/sec155.0 tokens/s
Time to First Token2.74s
Time to Answer2.74s
Available Providers
(LS internal units)| Provider | Input Price | Output Price |
|---|---|---|
| OpenAI | 200K | 1.3M |