Groq: Models Intelligence, Performance & Price
Analysis of Groq's models across key metrics including quality, price, output speed, latency, context window & more. This analysis is intended to support you in choosing the best model provided by Groq for your use-case. For more details including relating to our methodology, see our FAQs. Models analyzed: Llama 3.3 70B (Spec decoding), Llama 3.3 70B, Llama 3.2 90B (Vision), Llama 3.2 11B (Vision), Llama 3.1 8B, Llama 3.2 3B, Llama 3.2 1B, Gemma 2 9B, Mixtral 8x7B, DeepSeek R1 Distill Llama 70B, DeepSeek R1 Distill Llama 70B (Spec decoding), DeepSeek R1 Distill Qwen 32B, Qwen2.5 Coder 32B, Qwen2.5 Instruct 32B, Llama 3 70B, and Llama 3 8B.
Link:
Groq Model Comparison Summary
Intelligence:
DeepSeek R1 Distill Qwen 32BĀ andĀ DeepSeek R1 Distill Llama 70BĀ are the highest quality models offered by Groq, followed by DeepSeek R1 Distill Llama 70B (Spec decoding), 
Llama 3.3 70B (Spec decoding) & Llama 3.3 70B.Output Speed (tokens/s): Llama 3.2 1B (3181 t/s)Ā andĀ DeepSeek R1 Distill Llama 70B (Spec decoding) (1613 t/s)Ā are the fastest models offered by Groq, followed by Llama 3.2 3B, Llama 3.3 70B (Spec decoding) & Llama 3 8B.Latency (seconds):
Gemma 2 9B (0.23s)Ā and Ā Llama 3 70B (0.23s)Ā are the lowest latency models offered by Groq, followed by 
Mixtral 8x7B, Llama 3.2 11B (Vision) & Llama 3.1 8B.Blended Price ($/M tokens): Llama 3.2 1B ($0.04)Ā andĀ Llama 3.1 8B ($0.06)Ā are the cheapest models offered by Groq, followed by Llama 3 8B, Llama 3.2 3B & Llama 3.2 11B (Vision).Context Window Size:
Qwen2.5 Coder 32B (131k)Ā andĀ DeepSeek R1 Distill Qwen 32B (128k)Ā are the largest context window models offered by Groq, followed by DeepSeek R1 Distill Llama 70B, DeepSeek R1 Distill Llama 70B (Spec decoding) & Llama 3.3 70B.
DeepSeek R1 Distill Qwen 32BĀ andĀ DeepSeek R1 Distill Llama 70BĀ are the highest quality models offered by Groq, followed by DeepSeek R1 Distill Llama 70B (Spec decoding), Llama 3.3 70B (Spec decoding) & Llama 3.3 70B.Output Speed (tokens/s): Llama 3.2 1B (3181 t/s)Ā andĀ DeepSeek R1 Distill Llama 70B (Spec decoding) (1613 t/s)Ā are the fastest models offered by Groq, followed by Llama 3.2 3B, Llama 3.3 70B (Spec decoding) & Llama 3 8B.Latency (seconds): Gemma 2 9B (0.23s)Ā and Ā Llama 3 70B (0.23s)Ā are the lowest latency models offered by Groq, followed by Mixtral 8x7B, Llama 3.2 11B (Vision) & Llama 3.1 8B.Blended Price ($/M tokens): Llama 3.2 1B ($0.04)Ā andĀ Llama 3.1 8B ($0.06)Ā are the cheapest models offered by Groq, followed by Llama 3 8B, Llama 3.2 3B & Llama 3.2 11B (Vision).Context Window Size: Qwen2.5 Coder 32B (131k)Ā andĀ DeepSeek R1 Distill Qwen 32B (128k)Ā are the largest context window models offered by Groq, followed by DeepSeek R1 Distill Llama 70B, DeepSeek R1 Distill Llama 70B (Spec decoding) & Llama 3.3 70B.Highlights
Intelligence
Artificial Analysis Intelligence Index; Higher is better
Speed
Output Tokens per Second; Higher is better
Price
USD per 1M Tokens; Lower is better
Parallel Queries:
Prompt Length:
Features | Model Intelligence | Price | Output tokens/s | Latency | |||
|---|---|---|---|---|---|---|---|
Further Analysis | |||||||
 | DeepSeek R1 Distill Qwen 32B | 128k | 46 | $0.69 | 103.1 | 2.16 | |
| DeepSeek R1 Distill Llama 70B | 128k | 45 | $0.81 | 256.6 | 3.35 | |
| DeepSeek R1 Distill Llama 70B (Spec decoding) | 128k | 45 | $0.81 | 1,613.5 | 1.24 | |
| Llama 3.3 70B (Spec decoding) | 8k | 41 | $0.69 | 1,514.8 | 0.41 | |
| Llama 3.3 70B | 128k | 41 | $0.64 | 275.3 | 0.36 | |
| Qwen2.5 Instruct 32B | 128k | 37 | $0.79 | 198.1 | 0.35 | |
| Qwen2.5 Coder 32B | 131k | 36 | $0.79 | 384.2 | 0.35 | |
| Llama 3.2 90B (Vision) | 8k | 33 | $0.90 | 264.2 | 0.32 | |
| Llama 3 70B | 8k | 27 | $0.64 | 345.9 | 0.23 | |
| Llama 3.1 8B | 128k | 24 | $0.06 | 750.6 | 0.32 | |
| Llama 3.2 11B (Vision) | 8k | 22 | $0.18 | 749.8 | 0.32 | |
| Llama 3.2 3B | 8k | $0.06 | 1,546.2 | 0.34 | ||
| Llama 3.2 1B | 8k | $0.04 | 3,181.0 | 0.50 | ||
| Gemma 2 9B | 8k | $0.20 | 663.6 | 0.23 | ||
| Mixtral 8x7B | 33k | $0.24 | 566.9 | 0.24 | ||
| Llama 3 8B | 8k | $0.06 | 1,197.9 | 0.34 |
Key definitions
Context window: Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).
Output Speed: Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Latency: Time to first token of tokens received, in seconds, after API request sent. For models which do not support streaming, this represents time to receive the completion.
Price: Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Output Price: Price per token generated by the model (received from the API), represented as USD per million Tokens.
Input Price: Price per token included in the request/message sent to the API, represented as USD per million Tokens.
Time period: Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.