Groq develops high-performance AI accelerators designed to efficiently handle deep learning workloads and optimize computational tasks.
Pricing Change History
9 changesRemoved Llama 4 Scout (17Bx16E) and Qwen3 32B from the main LLM pricing table, and Qwen3-VL 32B from the Enterprise-only table; Minimax M2.5 updated to Minimax M2.7 in the Enterprise-only table.
New Enterprise-only LLM tier introduced with Minimax M2.5 and Qwen3-VL 32B models, priced as contact sales (no public per-token rate) Added Qwen 3.6 27B 131k model to on-demand LLM pricing at $0.60/1M input tokens, $3.00/1M output tokens (500 TPS)
GPT OSS 120B output token price decreased from $0.75 to $0.60 per million tokens (20% decrease) GPT OSS Safeguard 20B content moderation model added with input price $0.075 and output price $0.30 per million tokens
Quarterly roll-up · includes Dec 29 weekly
Llama 3 70B 8k model removed (previously $0.59 input/$0.79 output per million tokens) Batch API discount increased from 25% to 50%
DeepSeek R1 Distill Llama 70B speed increased from 275 to 400 tokens/sec 50% batch processing discount promotion ended (was doubled from 25% through end of April 2025)
Gemma 7B 8k Instruct model removed (was $0.07/M tokens for both input and output at 950 tokens/second) Llama 3 Groq 8B Tool Use Preview 8k model removed (was $0.19/M tokens for both input and output at 1250 tokens/second)
New Llama 3.3 70B SpecDec 8k model added with 1600 tokens/second speed, $0.59 input and $0.99 output pricing per million tokens Llama 3.3 70B Versatile speed increased from 250 to 275 tokens per second
Research Groq's pricing with the Pulse Agent
Ask anything about Groq's plans, packaging shifts, and how it compares across the market — answered live from verified data.
Ask the agentCurrent plans
Captured| Plan | Price |
|---|---|
| GPT OSS 20B | $0.08 per 1M input tokens + $0.3 per 1M output tokens |
| Whisper Large v3 (ASR) | $0.11 per hour transcribed |
| Whisper Large V3 Turbo (ASR) | $0.04 per hour transcribed |
| Canopy Labs Orpheus Arabic Saudi (TTS) | $40 per 1M characters |
| Canopy Labs Orpheus V1 English (TTS) | $22 per 1M characters |
| Llama 3.1 8B | Contact |
| Llama Prompt Guard 2 22M | $0.03 per 1M input tokens + $0.03 per 1M output tokens |
| Prompt Guard 2 86M | $0.04 per 1M output tokens + $0.04 per 1M input tokens |
| MiniMax M2.7 | Contact |
| Safety GPT OSS 20B | $0.3 per 1M output tokens + $0.08 per 1M input tokens |
| Qwen3.8-27B | $4 per 1M output tokens + $0.8 per 1M input tokens |
| Llama 3.3 70B | Contact |
| GPT OSS 120B | $0.15 per 1M input tokens + $0.6 per 1M output tokens |
Company Facts
How Groq compares with its peers
| Company | Entry price | Last change |
|---|---|---|
O OpenAI | $8 /mo | Oct 26, 2026 |
A Anthropic | $17 /mo, billed annually | Oct 12, 2026 |
G Google | Contact | Oct 12, 2026 |
S SambaNova | Free | Oct 26, 2026 |
L Lightning AI | $20 /mo, billed annually | Jul 13, 2026 |
H Huggingface | $9 /mo | Oct 16, 2026 |
Groq pricing FAQ
- How much does Groq cost?
- Groq prices on usage: the GPT OSS 20B plan is $0.08 per 1M input tokens + $0.3 per 1M output tokens.
- Does Groq have a free plan?
- No. Groq does not list a free tier.
- What pricing model does Groq use?
- Groq uses usage-based pricing across 13 plans.
- When did Groq last change its pricing?
- Sep 20, 2026: Groq drops Compound from Featured Models, leaving only GPT-OSS 120B.
- How often has Groq changed its pricing?
- Pulse has recorded 9 pricing changes for Groq since it started tracking on Nov 17, 2024; the pricing page was last captured on Sep 20, 2026.
- Who are Groq's pricing peers?
- Pulse compares Groq with OpenAI, Anthropic, Google, SambaNova, Lightning AI and Huggingface.