GPT OSS 120B
openai/gpt-oss-120b- OpenAI
- Family · GPT OSS
- Released 5 Aug 2025
- Open weights
- Text → Text
- Reasoning · adjustable
- Tools · JSON
Spec & list price
Context128K
Max output64K
Input / 1M$0.02
Output / 1M$0.063.6× input
Scores · standing
Intelligence11.6#223/489
Coding30.4#111/196
Usage
Tokens / day42.8B▼ 19% 7d
Usage rank#51
Requests / day8.4M
Tool calls / req0.04
Per request · 30 days
Avg input3.5K0.2× median
Avg output5170.7× median
Cache hit39%0.6× median
Reasoning share55%about median
Tool-call failures8.4%4.4× median
Consumption
Peak 108.0B 29 SepLow 37.4B 31 OctRank today #51
News & trends
4 mentions this week ▲ 100%this model · videos, posts, articles · 30 days
Trending · 14 days
- PROVEAI AI BOX 🤖 GPT-OSS 120B, управление мышью! #AI #ESP32 #Shorts1.1K views
- GPT-OSS-120B - Inside a Transformer8 views
News · latest from OpenAI
- GPT-6.1 Terrasimonwillison.net · 31 Oct
- GPT-5.6 Sol can beat Factorio on peaceful moderyanmadden.net · 31 Oct
- Post-training 4B model with RL to fly Tesla Roadster 21% better than GPT-6 Astraskypilot.ai · 30 Oct
- DeepSeek, Moonshot used OpenAI's GPT-6 Astra to train latest modelsreuters.com · 29 Oct
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 26 providers, latest day
Output speed#56/333113 tok/s1.7× faster than median
Fastest provider652 tok/sbest of 26 providers
First token#420.43 s2.4× quicker than median
First answer token#398.22 sthinking included
Best value in its score band
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases10.0
Technical details
- Reasoning
- on · medium
- Tokenizer
- GPT
- Knowledge cutoff
- 2024-06-30
- License
- Apache 2.0
- Parameters
- 117B
- Moderation
- none
- Downloads · likes
- 2.6M · 5.5K
Reasoning effort
- high
- medium · default
- low
Request parameters 21
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- temperature
- top_p
- top_k
+11 more
- min_p
- max_tokens
- stop
- seed
- frequency_penalty
- logit_bias
- logprobs
- presence_penalty
- repetition_penalty
- top_a
- top_logprobs
When to use
Best for
- Self-hosted reasoning and coding assistants
- Math and science problem solving
- Tool-calling agents with structured output
- Classification and extraction with adjustable reasoning effort
Think twice for
- Image or audio input
- Tasks needing the very longest context
- Newest-generation agentic work
Research papers
- Discovering Jailbreaks via Curriculum of Models28 Oct
- How Much Does the Hint Matter? Scope Conditions for Hindsight Self-Distillation in Financial Reasoning26 Oct
- SEMA-Auto: Sighted and Efficient Multi-Turn Attack with Autonomy26 Oct
- Local Work, Global Progress: A Convergence Theory of AI Agent Collaboration20 Oct
- Multi-Agentic AI for Intent-Driven and Conflict-Aware rApp Policy Orchestration in Open RAN17 Oct
- Plan or Retrieve First? A Scale Study of Tool Discovery Architectures13 Oct
- Beyond the Verdict: Agentic Code Judging with Verifiable Feedback13 Oct
- MagicBeanStalk: Escaping the RLVR Zero Advantage Trap with Recursive Data Synthesis8 Oct
Social
Most watched and discussed · 90 days
- PROVEAI AI BOX 🤖 GPT-OSS 120B, управление мышью! #AI #ESP32 #Shortsinjener_khrenov · 1.1K views · 31 Oct
- GPT-OSS-120B - Inside a TransformerThe Residual Stream · 8 views · 23 Oct
Same family
GPT OSS- 10.0
Alternatives
other vendors · similar capability- 12.3
- 8.4
- 13.1
- 10.8
Using GPT OSS 120B at volume?
Sign in to ask for a volume discount. You will continue in the OSMEXO dashboard.