Skip to content

GPT-OSS 20B - Free Lightweight Model on Groq, 131K Context

Model ID openai/gpt-oss-20b
Available directly on GroqCloud’s free plan with no credit card required. Rate limits: 30 RPM / 1,000 RPD / 8K TPM / 200K TPD.

Model Information

AttributeValue
Context length131,072 tokens
Max output65,536 tokens
Input price$0 (free plan)
Output price$0 (free plan)
ProviderOpenAI / Groq
StatusAvailable on free plan
Input modalitiesText
Inference speed~1,000 tokens/s

Description

GPT-OSS 20B is the lightweight member of the GPT-OSS family, with 20B parameters and an official speed of about 1,000 tokens/s. It keeps reasoning, function calling, and structured output capabilities, making it suitable for high-frequency calls, real-time interaction, and lightweight agents.

On Groq’s free plan, the 20B and 120B share the same rate limits (30 RPM / 1,000 RPD / 8K TPM / 200K TPD), so it makes sense to route simple tasks to the 20B and leave complex ones to the 120B.

Usage Example

curl https://api.groq.com/openai/v1/chat/completions \
  -H "Authorization: Bearer $GROQ_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai/gpt-oss-20b",
    "messages": [{"role": "user", "content": "Condense this paragraph into one sentence"}]
  }'

Related Links

📚 Related Free Models

These models are also available for $0 on Groq. Choose by context length and modality.

ModelContextDescription
GPT-OSS 120B131KOpenAI’s 120B open-source reasoning model excels at
Qwen 3.8 27B131KAlibaba’s 27B multimodal model supports image input,

📋 View all Groq free models →

🔗 Site Navigation

Last updated on · Source: Groq official documentation and pricing pages. Quotas and prices may change at any time.