OpenAI

GPT OSS 120B

Type
language model
Context
131K tokens
Max output
33K
Released
5 August 2025
API string
accounts/fireworks/models/gpt-oss-120b
Weights
open

Prices by provider

Provider Input, $ per 1M Output, $ per 1M In our data since
Weights & Biases $0.03 $0.17 31 Jul 2026
LLM Gateway $0.032 $0.14 31 Jul 2026
DeepInfra openai/gpt-oss-120b $0.037 $0.17 31 Jul 2026
OpenRouter openai/gpt-oss-120b $0.037 $0.17 31 Jul 2026
Kilo Code $0.039 $0.19 31 Jul 2026
tensorx $0.04 $0.2 2 Aug 2026
Helicone $0.04 $0.16 31 Jul 2026
io.net $0.04 $0.4 31 Jul 2026
novita-ai $0.05 $0.25 31 Jul 2026
DeepInfra deepinfra/openai/gpt-oss-120b $0.05 $0.45 31 Jul 2026
NanoGPT openai/gpt-oss-120b $0.05 $0.25 31 Jul 2026
Novita AI $0.05 $0.25 31 Jul 2026
SiliconFlow $0.05 $0.45 31 Jul 2026
dinference $0.068 $0.27 31 Jul 2026
Venice AI $0.07 $0.3 1 Aug 2026
neon $0.072 $0.28 31 Jul 2026
Databricks databricks-gpt-oss-120b $0.072 $0.28 31 Jul 2026
Abacus.AI $0.08 $0.44 31 Jul 2026
OVHCloud AI Endpoints ovhcloud/gpt-oss-120b $0.08 $0.4 31 Jul 2026
Google Vertex AI openai/gpt-oss-120b-maas $0.09 $0.36 31 Jul 2026
Merge Gateway $0.09 $0.36 31 Jul 2026
OVHCloud AI Endpoints gpt-oss-120b $0.09 $0.47 31 Jul 2026
submodel $0.1 $0.5 31 Jul 2026
synthetic $0.1 $0.1 31 Jul 2026
Baseten $0.1 $0.5 31 Jul 2026
DigitalOcean Gradient AI $0.1 $0.7 1 Aug 2026
Vercel $0.1 $0.5 31 Jul 2026
frogbot $0.15 $0.6 31 Jul 2026
aiand $0.15 $0.6 31 Jul 2026
aki-io $0.15 $0.55 31 Jul 2026
tinfoil $0.15 $0.6 31 Jul 2026
bedrock_mantle $0.15 $0.6 31 Jul 2026
tensormesh $0.15 $0.6 31 Jul 2026
Amazon Bedrock $0.15 $0.6 31 Jul 2026
FastRouter $0.15 $0.6 31 Jul 2026
Fireworks AI $0.15 $0.6 31 Jul 2026
Google Vertex AI vertex_ai/openai/gpt-oss-120b-maas $0.15 $0.6 31 Jul 2026
Groq $0.15 $0.6 31 Jul 2026
IBM watsonx.ai $0.15 $0.6 31 Jul 2026
Microsoft Foundry $0.15 $0.6 31 Jul 2026
NEAR AI $0.15 $0.55 31 Jul 2026
Nebius Token Factory $0.15 $0.6 31 Jul 2026
Pioneer $0.15 $0.6 31 Jul 2026
Scaleway $0.15 $0.6 31 Jul 2026
Together AI $0.15 $0.6 31 Jul 2026
Databricks databricks/databricks-gpt-oss-120b $0.15 $0.6 31 Jul 2026
OpenRouter openrouter/openai/gpt-oss-120b $0.18 $0.8 31 Jul 2026
Replicate $0.18 $0.72 31 Jul 2026
hyper $0.188 $0.7 1 Aug 2026
berget $0.22 $0.83 31 Jul 2026
SambaNova $0.22 $0.59 31 Jul 2026
greenpt $0.228 $0.798 31 Jul 2026
evroc $0.23 $0.92 31 Jul 2026
Hugging Face $0.25 $0.69 31 Jul 2026
cloudflare-workers-ai $0.35 $0.75 31 Jul 2026
Cerebras $0.35 $0.75 31 Jul 2026
Cloudflare AI Gateway $0.35 $0.75 31 Jul 2026
Cloudflare Workers AI $0.35 $0.75 31 Jul 2026
stackit $0.53 $0.76 31 Jul 2026
crusoe $0.8 $0.8 31 Jul 2026
regolo-ai $1 $4.2 31 Jul 2026
NanoGPT TEE/gpt-oss-120b $2 $2 31 Jul 2026
cloudferro-sherlock $2.92 $2.92 31 Jul 2026

Only the base tier and only per-token prices are shown. Batch, discounted and cached rates, as well as prices per image or per second of video, do not go into this table: they cannot stand in the same column as a price per million tokens.

A single provider appears several times if it sells this model under different identifiers and at different prices — for example, at different weight precision. The identifier is shown next to the name. Identical offers that arrived from two sources under different spellings are merged into one row.

The date in the last column is the day this price first entered our collection. The price may well be older: before that day we simply were not recording it. It has not changed since — otherwise a new row with a new date would stand in its place.

Measurement results

Task set Result Run conditions Measured by
Arena Score, mathematics 1,388.15 1,374–1,402
Arena Score in Spanish 1,384.44 1,362–1,407
Arena Score, programming 1,380.59 1,373–1,388
Arena Score in English 1,373.99 1,368–1,380
Arena Score in French 1,372.17 1,343–1,402
Arena Score, overall 1,365.68 1,361–1,370
Arena Score, hard prompts 1,363.95 1,358–1,370
Arena Score, expert questions 1,357.7 1,341–1,374
Arena Score, multi-turn dialogue 1,340.1 1,331–1,349
Arena Score in Russian 1,338.46 1,324–1,353
Arena Score, instruction following 1,319.52 1,312–1,327
Arena Score, long queries 1,319.16 1,311–1,327
Arena Score, creative writing 1,273.41 1,264–1,283
Mock AIME 2024–2025 — olympiad problems 88.88 % effort: high · with a tuned harness Epoch evaluations
Creative writing (Lech Mazur’s evaluation) 77.1 % lechmazur/writing Github repository
GPQA Diamond — graduate-level questions 67.68 % effort: high · with a tuned harness Epoch evaluations
WeirdML — unusual machine learning tasks 48.17 % effort: high WeirdML Leaderboard
Fiction.LiveBench — holding a long context 44.4 % Fiction.live leaderboard
Aider Polyglot — code edits in six languages 41.8 % effort: high · with a tuned harness Aider LLM Leaderboards
Terminal-Bench — working in the command line 18.7 % · with a tuned harness Terminal-Bench v2 Leaderboard
Chess puzzles 15.83 % effort: high Epoch evaluations
SimpleQA Verified — factual accuracy 13.9 % effort: high Epoch evaluations
SimpleBench — trick questions 6.52 % SimpleBench Leaderboard
APEX-Agents 4.7 %
CritPt — physics problems 1.14 % effort: high