Skip to content

PricingSaaS joins Willingness to PayRead the announcement →

PricingSaaS
Sign in
HomeGroqAll changes
Packaging change·Nov 17 → Jan 3, 2025·Dec 2024·AI & Machine Learning

Groq Introduces New Llama 3.3 70B Model with Competitive Pricing and High Speed of 1600 Tokens/Second

GroqPackagingNew planFeature addedLimit increasePlan renamed

What changed

New plan

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

Feature added

Vision model billing clarification added: images are billed at 6,400 tokens per image

Limit increase

Llama 3.3 70B Versatile speed increased from 250 to 275 tokens per second

Plan renamed

Llama 3.1 70B Versatile model renamed to Llama 3.3 70B Versatile

Before and after

Nov 17, 2024 vs Jan 3, 2025
Pulse.
100%
Change 1 of 4 · plan added

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

Sign in to view the full comparison

See the before/after screenshots and every marked-up change.

Sign up freeLog in

More from Groq

Full history →
  1. Week 30, 2026·packaging
    Groq drops Llama 4 Scout and Qwen3 32B, upgrades Minimax to M2.7
  2. Q2 2026·packaging
    Groq: Qwen 3.6 27B added, new Enterprise-only LLM tier + 1 more change
  3. Q4 2025·decrease
    Groq Reduces GPT OSS 20B Input and Output Token Prices by 25% and 40%, Enhancing Cost Efficiency for Users
  4. Week 1, 2026·packaging
    Groq swaps PlayAI for Orpheus TTS, cuts text-to-speech price 56%
  5. Q3 2025·improved
    Groq Removes Multiple Models While Adding High-Performance GPT OSS 20B and 120B with New Prompt Caching Feature