OpenAI: GPT-5 Nano

openai/gpt-5-nano

GPT-5 Nano is OpenAI's budget option at $0.05/$0.4 per million input/output tokens, with a 400K-token context. It sits mid-board on capability, at #144 of 234 on the TryAii Score — it holds up best on instruction-following (#2 on IFEval). It's among the fastest models we track, at roughly 77-95 tokens/sec.

Prompt $/1M

$0.050

Completion $/1M

$0.400

Context

400,000 tok

Best provider

104 tok/s · OpenAI

Specs

Model id: openai/gpt-5-nano
Display name: OpenAI: GPT-5 Nano
Modality: text+image+file->text
Tokenizer: GPT
Instruct type: —
Context length: 400,000
Max completion: 128,000
Prompt $/1M: $0.050
Completion $/1M: $0.400
Free tier: no

Capabilities

Function callingyes
JSON modeno
System promptno
Visionno
Moderatedyes

Benchmark scores

Each axis is one benchmark, normalized to the best model in the field.

No benchmark data for this model.

Speed by provider

Provider	tok/s	TTFT	Total latency
OpenAIfastest	104	2,550 ms	3,357 ms
Azure	68	2,662 ms	3,023 ms
Artificial Analysis	0	0 ms	—

Prompt caching by provider

Provider	Min tokens to cache	Notes
OpenAI	1,024 tokens	GPT-5.x pre-5.6: 90% cached-read discount, free cache writes. 5-10 min sliding TTL; extended retention up to 24h on eligible models (non-ZDR default since May 2026).