Groq models and pricing
We track 11 models on Groq from 5 labs. The cheapest paid model is Llama Prompt Guard 2 22M, at $0.03 per million input tokens and $0.03 per million output tokens. Input prices range from $0.03 to $0.80 and output prices from $0.03 to $4 per million tokens. 1 model is listed free. The largest context window is 131K tokens (Qwen3.8 27B, GPT OSS Safeguard 20B and 2 other models). Prices last verified 18 Nov 2026.
Prices in USD per million tokens. Official site
Cheapest models on Groq
- Llama Prompt Guard 2 22M$0.03 / $0.03
- Prompt Guard 2 86M$0.04 / $0.04
- GPT OSS Safeguard 20B$0.07 / $0.30
- gpt-oss-20b$0.07 / $0.30
- gpt-oss-120b$0.15 / $0.60
Input / output per million tokens, paid models, by combined price.
Newest models on Groq
- Qwen3.8 27B14 Aug 2026
- Canopy Labs Orpheus V1 English19 Dec 2025
- Canopy Labs Orpheus Arabic Saudi16 Dec 2025
- GPT OSS Safeguard 20B29 Oct 2025
- gpt-oss-20b5 Aug 2025
By the lab's release date.
All Groq models and prices
| Model | Lab | Input | Output | Context | Source |
|---|---|---|---|---|---|
| Qwen3.8 27B qwen/qwen3.8-27b | Alibaba (Qwen) | $0.80 | $4 | 131K | models.dev |
| Canopy Labs Orpheus V1 English canopylabs/orpheus-v1-english | Canopylabs | – | – | 4K | models.dev |
| Canopy Labs Orpheus Arabic Saudi canopylabs/orpheus-arabic-saudi | Canopylabs | – | – | 4K | models.dev |
| GPT OSS Safeguard 20B openai/gpt-oss-safeguard-20b | OpenAI | $0.07 | $0.30 | 131K | models.dev |
| gpt-oss-20b openai/gpt-oss-20b | OpenAI | $0.07 | $0.30 | 131K | models.dev |
| gpt-oss-120b openai/gpt-oss-120b | OpenAI | $0.15 | $0.60 | 131K | models.dev |
| Prompt Guard 2 86M meta-llama/llama-prompt-guard-2-86m | Meta | $0.04 | $0.04 | 512 | models.dev |
| Llama Prompt Guard 2 22M meta-llama/llama-prompt-guard-2-22m | Meta | $0.03 | $0.03 | 512 | models.dev |
| ALLaM-2-7b allam-2-7b | Sdaia | Free | Free | 4K | models.dev |
| Whisper Large V3 Turbo whisper-large-v3-turbo | OpenAI | – | – | 448 | models.dev |
| Whisper Large v3 whisper-large-v3 | OpenAI | – | – | 448 | models.dev |
Frequently asked questions
How much does Groq cost per million tokens?
Prices on Groq depend on the model. Across the 11 models we track, input costs $0.03 to $0.80 per million tokens and output costs $0.03 to $4 per million tokens. 1 model is listed free. Prices last verified 18 Nov 2026.
What is the cheapest model on Groq?
Llama Prompt Guard 2 22M from Meta is the cheapest paid model we track on Groq, at $0.03 per million input tokens and $0.03 per million output tokens. Next are Prompt Guard 2 86M ($0.04 / $0.04) and GPT OSS Safeguard 20B ($0.07 / $0.30). Prices last verified 18 Nov 2026.
How many models does Groq offer?
We currently track 11 active models on Groq from 5 labs, led by OpenAI (5), Canopylabs (2), Meta (2) and Alibaba (Qwen) (1).
Which model on Groq has the largest context window?
Qwen3.8 27B, GPT OSS Safeguard 20B and 2 other models share the largest context window we track on Groq: 131,072 tokens (131K).