OpenAI: gpt-oss-120b

openai/gpt-oss-120b

gpt-oss-120b is an open-weight, 117B-parameter Mixture-of-Experts (MoE) language model from OpenAI designed for high-reasoning, agentic, and general-purpose production use cases. It activates 5.1B parameters per forward pass and is optimized...

Prompt $/1M

$0.037

Completion $/1M

$0.170

Context

131,072 tok

Best provider

573 tok/s · Cerebras

Specs

Model id: openai/gpt-oss-120b
Display name: OpenAI: gpt-oss-120b
Modality: text->text
Tokenizer: GPT
Instruct type: —
Context length: 131,072
Max completion: 131,072
Prompt $/1M: $0.037
Completion $/1M: $0.170
Free tier: no

Capabilities

Function callingno
JSON modeno
System promptno
Visionno
Moderatedno

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
Cerebrasfastest	573	249 ms	322 ms
Artificial Analysis	343	452 ms	—
Groq	331	225 ms	343 ms
Amazon Bedrock	283	2,126 ms	2,953 ms
SambaNova	231	1,300 ms	1,654 ms
Nebius	153	500 ms	802 ms
DeepInfra	152	484 ms	560 ms
Fireworks	126	346 ms	424 ms
Mara	117	4,346 ms	5,586 ms
AtlasCloud	112	595 ms	624 ms
Parasail	86	498 ms	599 ms
Ambient	72	586 ms	741 ms
Together	70	289 ms	354 ms
Novita	57	649 ms	896 ms
Io Net	53	1,225 ms	1,726 ms
Phala	46	1,321 ms	1,519 ms
Google	44	1,236 ms	1,811 ms
Clarifai	40	1,363 ms	2,820 ms
OpenInference	36	471 ms	581 ms
BaseTen	31	470 ms	566 ms
WandB	25	565 ms	646 ms
Mancer 2	24	1,990 ms	2,427 ms
DekaLLM	21	1,432 ms	1,711 ms
DigitalOcean	18	1,634 ms	1,927 ms
SiliconFlow	10	6,015 ms	10,434 ms
Chutes	5	2,676 ms	2,782 ms