Groq Introduces New High-Performance Models and 50% Discount on Batch API for Paid GroqCloud Customers
What happened
Between the Jan 3, 2025 and Apr 1, 2025 captures of Groq's pricing page, Pulse recorded 12 changes: Llama 3 Groq 8B Tool Use Preview 8k model removed (was $0.19/M tokens for both input and output at 1250 tokens/second); Qwen QwQ 32B (Preview) 128k model added with input price $0.29/M tokens and output price $0.39/M tokens at 400 tokens/second; Llama 3 Groq 70B Tool Use Preview 8k model removed (was $0.89/M tokens for both input and output at 335 tokens/second); Qwen 2.5 Coder 32B Instruct 128k model added with input price $0.79/M tokens and output price $0.79/M tokens at 390 tokens/second; DeepSeek R1 Distill Llama 70B model added with input price $0.75/M tokens and output price $0.99/M tokens at 275 tokens/second; Qwen 2.5 32B Instruct 128k model added with input price $0.79/M tokens and output price $0.79/M tokens at 200 tokens/second; Mistral Saba 24B model added with input price $0.79/M tokens and output price $0.79/M tokens at 330 tokens/second; Batch API introduced with 25% discount rate, doubled to 50% off through end of April 2025 for paid GroqCloud customers; Text-to-Speech (TTS) Models section introduced with PlayAI Dialog v1.0 model priced at $50.00 per million characters at 140 characters/second; Mixtral 8x7B Instruct 32k model removed (was $0.24/M tokens for both input and output at 575 tokens/second); DeepSeek R1 Distill Qwen 32B 128k model added with input price $0.69/M tokens and output price $0.69/M tokens at 140 tokens/second; Gemma 7B 8k Instruct model removed (was $0.07/M tokens for both input and output at 950 tokens/second). Pulse files it as a pricing change (plan removed, new plan, discount added, feature added) with a downward direction. Pulse's latest capture (Sep 20, 2026) lists GPT OSS 20B at $0.08 per 1M input tokens + $0.3 per 1M output tokens, Whisper Large v3 (ASR) at $0.11 per hour transcribed, Whisper Large V3 Turbo (ASR) at $0.04 per hour transcribed, Canopy Labs Orpheus Arabic Saudi (TTS) at $40 per 1M characters and Canopy Labs Orpheus V1 English (TTS) at $22 per 1M characters.
Groq's previous recorded change was "Groq Introduces New Llama 3.3 70B Model with Competitive Pricing and High Speed of 1600 Tokens/Second" (Dec 2024). Groq has been tracked since 2024-11-17.
In the same window Anthropic also changed their AI & Machine Learning pricing pages. Those moves are listed further down, and the Pulse Agent can compare them side by side.
“GroqCloud” is the name Groq’s pricing page uses; Pulse files the change under Groq, the company it tracks.
- — · PlanPlan removed
Llama 3 Groq 8B Tool Use Preview 8k model removed (was $0.19/M tokens for both input and output at 1250 tokens/second)
- — · PlanNew planNot listed400 tokens/second
- — · PlanPlan removed
Llama 3 Groq 70B Tool Use Preview 8k model removed (was $0.89/M tokens for both input and output at 335 tokens/second)
- — · PlanAdded
BeforeNot listed
AfterQwen 2.5 Coder 32B Instruct 128k model
- — · PlanAdded
BeforeNot listed
AfterDeepSeek R1 Distill Llama 70B model
- — · PlanAdded
BeforeNot listed
AfterQwen 2.5 32B Instruct 128k model
- — · PlanAdded
BeforeNo Mistral Saba 24B model
AfterMistral Saba 24B model
- — · DiscountAddedNo Batch APIBatch API
- — · FeatureFeature addedNot listed$50.00 per million
- — · PlanPlan removed
Mixtral 8x7B Instruct 32k model removed (was $0.24/M tokens for both input and output at 575 tokens/second)
- — · PlanAdded
BeforeNot listed
AfterDeepSeek R1 Distill Qwen 32B 128k model
- — · PlanPlan removed
Gemma 7B 8k Instruct model removed (was $0.07/M tokens for both input and output at 950 tokens/second)
| Plan | Field | Before | After | Change |
|---|---|---|---|---|
| — | Plan | — | 1250 tokens/second | Plan removed |
| — | Plan | Not listed | 400 tokens/second | New plan |
| — | Plan | — | 335 tokens/second | Plan removed |
| — | Plan | Not listed | Qwen 2.5 Coder 32B Instruct 128k model | Added |
| — | Plan | Not listed | DeepSeek R1 Distill Llama 70B model | Added |
| — | Plan | Not listed | Qwen 2.5 32B Instruct 128k model | Added |
| — | Plan | No Mistral Saba 24B model | Mistral Saba 24B model | Added |
| — | Discount | No Batch API | Batch API | Added |
| — | Feature | Not listed | $50.00 per million | Feature added |
| — | Plan | — | 575 tokens/second | Plan removed |
| — | Plan | Not listed | DeepSeek R1 Distill Qwen 32B 128k model | Added |
| — | Plan | — | 950 tokens/second | Plan removed |
Values as written on Groq's public pricing page in the two captures Pulse compared; where the page states no number, the cells give the wording before and after.
Evidence
The two captures Pulse comparedFull-page captures of Groq's pricing page (groq.com) from Jan 3, 2025 and Apr 1, 2025; the change is marked on the page, where "Not listed" becomes "400 tokens/second".
Sign in to see the evidence
The two full-page captures with this change marked up are free with an account.
Groq, the last 90 days
Full history- Week 1, 2026packagingGroq swaps PlayAI for Orpheus TTS, cuts text-to-speech price 56%
- Q3 2025packagingGroq Removes Multiple Models While Adding High-Performance GPT OSS 20B and 120B with New Prompt Caching Feature
- Q2 2025improvedGroq Introduces Llama 4 Maverick and Qwen3 Models Amid Removal of Multiple High-Cost Plans and End of 50% Batch Processing Discount
- Q1 2025packagingGroq Introduces New High-Performance Models and 50% Discount on Batch API for Paid GroqCloud Customers · this change
- Dec 2024packagingGroq Introduces New Llama 3.3 70B Model with Competitive Pricing and High Speed of 1600 Tokens/Second
Across AI & Machine Learning in the same window
All AI & Machine Learning changes- AnthropicJan 10Anthropic API Introduces Tiered Pricing and Prompt Caching for Sonnet 4.5, Restructuring Costs Based on Token Length
- AnthropicJan 10Anthropic API Introduces Code Execution Tool with 50 Free Hours Daily, Enhancing Service Offerings and Flexibility for Users
- AnthropicJan 10Anthropic API Cuts Flagship Opus Model Prices by 67% in Major Update to Version 4.5, Launching New Pricing Structure
- AnthropicJan 10Anthropic Max Plan Now Includes Memory Across Conversations, Enhancing User Experience and Functionality
- AnthropicJan 10Anthropic Introduces Max Plan at $100+/Month, Offering 5x-20x More Usage Than Pro Plan
Questions about this change
Did Groq raise prices in April 2025?
No list-price increase was recorded between Jan 3, 2025 and Apr 1, 2025. The change Pulse detected was a pricing change: plan removed, new plan, discount added, feature added.
What changed on Groq's pricing page between Jan 3, 2025 and Apr 1, 2025?
Llama 3 Groq 8B Tool Use Preview 8k model removed (was $0.19/M tokens for both input and output at 1250 tokens/second); plan: Not listed → 400 tokens/second; Llama 3 Groq 70B Tool Use Preview 8k model removed (was $0.89/M tokens for both input and output at 335 tokens/second); plan: Not listed → Qwen 2.5 Coder 32B Instruct 128k model, added; plan: Not listed → DeepSeek R1 Distill Llama 70B model, added; plan: Not listed → Qwen 2.5 32B Instruct 128k model, added; plan: No Mistral Saba 24B model → Mistral Saba 24B model, added; discount: No Batch API → Batch API, added; feature: Not listed → $50.00 per million; Mixtral 8x7B Instruct 32k model removed (was $0.24/M tokens for both input and output at 575 tokens/second); plan: Not listed → DeepSeek R1 Distill Qwen 32B 128k model, added; Gemma 7B 8k Instruct model removed (was $0.07/M tokens for both input and output at 950 tokens/second). Pulse compared the captures it took on Jan 3, 2025 and Apr 1, 2025; the before/after table above lists each item as written on the page.
How often does Groq change its pricing?
Pulse recorded 1 other pricing-page change for Groq in the 90 days before this one; the previous change was "Groq Introduces New Llama 3.3 70B Model with Competitive Pricing and High Speed of 1600 Tokens/Second" in Dec 2024; Groq has been tracked since 2024-11-17.