Model specification · 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...

SpecificationsReported values
Context window
131K
Maximum output
66K
Architecture
text->text
Tokenizer
GPT
Knowledge cutoff
2024-06-30
Moderated
No
Input modalities
text
Output modalities
text
Full pricingUSD where reported
PriceStandard
Input$0.150 / M
Output$0.600 / M
Benchmark indexesReported indexes
Intelligence
12.3
Coding
30.4
Agentic
6.2
Design Arena Elo
975
Example workload costsStandard pricing only
WorkloadInput / outputCache shareEstimated cost
Quick check10K / 1K × 10%$0.00210
Long analysis100K / 4K × 150%$0.014
Workday20K / 2K × 1,00030%$3.75
More from OpenAIAll OpenAI models
ModelContextInput / MOutput / MIntel.
GPT-6 Astra1.1M$10.00$50.0052.8
GPT-5.6 Sol1.1M$2.00$10.0047.1
GPT-5.6 Terra1.1M$2.00$12.0042.3
GPT-5.51.1M$5.00$30.0038.6
GPT-5.6 Luna1.1M$0.200$1.2037.5
GPT-5.4 Mini400K$0.750$4.5024.6
GPT-5.4 Nano400K$0.200$1.2521.2
GPT-5 Mini400K$0.250$2.0017.4
o3 Mini High200K$1.10$4.4011
gpt-oss-20b131K$0.030$0.1309