impulse

gpt-oss-120b

Open weights

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.

API id gpt-oss-120bReleased 5 Aug 2025 (Aug 2025)
Input / 1M
$0.03
at W&B Inference
Output / 1M
$0.17
at W&B Inference
Context window
131K
tokens
Max output
131K
tokens per reply
Hosts
28
54 listings
Released
Aug 2025
5 Aug 2025

Pricing

List prices at W&B Inference, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.03$0.17

What it costs a month

At W&B Inference list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out$21.75
  • RAG over documents2,000 requests a day, 8K in / 0.6K out$20.52
  • Coding agent1,000 requests a day, 40K in / 2K out$46.20
  • Overnight batch20,000 requests a day, 4K in / 0.4K out$113
Use your own numbers

Where to run it

28 hosts serve gpt-oss-120b. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
W&B InferenceInference
gpt-oss-120b
$0.03 / $0.17
DeepInfraInference
gpt-oss-120b
$0.037 / $0.17
+8% vs headline
gpt-oss-120b
$0.0473 / $0.23
+43% vs headline
GMI CloudInference
gpt-oss-120b
$0.05 / $0.25
+54% vs headline
Novita AIInference
gpt-oss-120b
$0.05 / $0.25
+54% vs headline
SailCloud
gpt-oss-120b
$0.06 / $0.40
+123% vs headline
gpt-oss-120b-maas
$0.09 / $0.36
+142% vs headline
gpt-oss-120b
$0.08 / $0.40
+146% vs headline
BasetenInference
gpt-oss-120b
$0.10 / $0.50
+208% vs headline
ai&Cloud
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
Retires 15 Sep 2027
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
ZeroGPUCloud
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
openai.gpt-oss-120b-1:0
$0.15 / $0.60
+304% vs headline
openai.gpt-oss-120b
$0.15 / $0.60
+304% vs headline
Fireworks AIInference
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
GroqInference
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
Together AIInference
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
databricks-gpt-oss-120b
$0.15 / $0.60
+304% vs headline
gpt-oss-120b
$0.159 / $0.636
+328% vs headline
SambaNovaInference
gpt-oss-120b
$0.22 / $0.59
+381% vs headline
ReplicateInference
gpt-oss-120b
$0.18 / $0.72
+385% vs headline
gpt-oss-120b
$0.20 / $0.80
+438% vs headline
gpt-oss-120b
$0.35 / $0.75
+592% vs headline
CerebrasInference
gpt-oss-120b
$0.35 / $0.75
+592% vs headline
CrusoeInference
gpt-oss-120b
$0.80 / $0.80
+1131% vs headline
gpt-oss-120b-mxfp-GGUF
-
Show 25 more listings: regions, snapshots and variants
Amazon Bedrock
Mantle
openai.gpt-oss-120b
$0.15 / $0.60
+304% vs headline
ap-southeast-2 · Mantle
openai.gpt-oss-120b
$0.154 / $0.618
+316% vs headline
ap-southeast-3 · Mantle
openai.gpt-oss-120b
$0.16 / $0.62
+323% vs headline
ap-southeast-3
openai.gpt-oss-120b-1:0
$0.16 / $0.62
+323% vs headline
eu-west-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.70
+377% vs headline
eu-west-1
openai.gpt-oss-120b-1:0
$0.18 / $0.70
+377% vs headline
ap-south-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.71
+381% vs headline
ap-south-1
openai.gpt-oss-120b-1:0
$0.18 / $0.71
+381% vs headline
us-gov-west-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.72
+385% vs headline
us-gov-east-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.72
+385% vs headline
us-gov-west-1
openai.gpt-oss-120b-1:0
$0.18 / $0.72
+385% vs headline
us-gov-east-1
openai.gpt-oss-120b-1:0
$0.18 / $0.72
+385% vs headline
us-gov
us-gov.openai.gpt-oss-120b-1:0
$0.18 / $0.72
+385% vs headline
ap-northeast-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.73
+388% vs headline
sa-east-1 · Mantle
openai.gpt-oss-120b
$0.18 / $0.73
+388% vs headline
ap-northeast-1
openai.gpt-oss-120b-1:0
$0.18 / $0.73
+388% vs headline
sa-east-1
openai.gpt-oss-120b-1:0
$0.18 / $0.73
+388% vs headline
eu-central-1 · Mantle
openai.gpt-oss-120b
$0.20 / $0.79
+435% vs headline
eu-south-1 · Mantle
openai.gpt-oss-120b
$0.20 / $0.79
+435% vs headline
eu-central-1
openai.gpt-oss-120b-1:0
$0.20 / $0.79
+435% vs headline
eu-south-1
openai.gpt-oss-120b-1:0
$0.20 / $0.79
+435% vs headline
eu-west-2 · Mantle
openai.gpt-oss-120b
$0.23 / $0.93
+523% vs headline
eu-west-2
openai.gpt-oss-120b-1:0
$0.23 / $0.93
+523% vs headline
Fireworks AI
Alias
gpt-oss-120b
$0.15 / $0.60
+304% vs headline
Inference APIs
Alias
gpt-oss-120b
$0.20 / $0.80
+438% vs headline

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
Text
Produces
Text
Open weights
License: Apache-2.0
Runs in about 280.4 GB of GPU memory

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than W&B Inference.

Smaller OpenAI models

No cheaper OpenAI model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • Qwen3 32B
    Alibaba Qwen · on Lambda AI
    $0.05 / $0.10
    Has every key capability
    4% less
  • Gemma 4 E2B
    Google · on Amazon Bedrock
    $0.04 / $0.08
    Has every key capability
    23% less
  • gpt-oss-20b
    OpenAI · on Darkbloom
    $0.0145 / $0.07
    Has every key capability
    56% less
  • Qwen3 8B
    Alibaba Qwen · on Novita AI
    $0.035 / $0.138
    Has every key capability
    7% less

Benchmarks

Scores published on the model card.

  • Aider Polyglot41.8%

Call it

Using W&B Inference's API.

Python
from litellm import completion

response = completion(
    model="wandb/gpt-oss-120b",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Quick answers

How much does gpt-oss-120b cost?

On W&B Inference, gpt-oss-120b costs $0.03 per million input tokens and $0.17 per million output tokens.

Which APIs offer gpt-oss-120b?

gpt-oss-120b runs on 28 APIs: W&B Inference, DeepInfra, Kyma API, GMI Cloud, Novita AI, Sail, Vertex AI, OVHcloud AI Endpoints, Baseten and 19 more.

What is gpt-oss-120b's context window?

131K tokens, with up to 131K tokens in one reply.

Is gpt-oss-120b being retired?

Azure AI Foundry retires it on 15 Sep 2027. 27 hosts keep serving it.

What are cheaper alternatives to gpt-oss-120b?

Models that cost less and keep most of what it can do: Qwen3 32B on Lambda AI (4% less). Savings are on a mix of 3 input tokens to 1 output token.

When was gpt-oss-120b released?

OpenAI released it on 5 Aug 2025.

What is gpt-oss-120b's API model id?

Use gpt-oss-120b with W&B Inference. Other hosts use their own ids: gpt-oss-120b-maas on Vertex AI, openai.gpt-oss-120b-1:0 on Amazon Bedrock, openai.gpt-oss-120b on Oracle OCI, databricks-gpt-oss-120b on Databricks.

Sources