Groq: Models Intelligence, Performance & Price
This analysis is intended to support you in choosing the best model provided by Groq for your use-case.
Most Intelligent
Intelligence Index
Total 6 models
Fastest
gpt-oss-20b (high)gpt-oss-20b (low)
Llama 3.1 8Bgpt-oss-120b (high)gpt-oss-120b (low)Output speed
Total 6 models
Lowest Price
Llama 3.1 8Bgpt-oss-20b (low)gpt-oss-20b (high)gpt-oss-120b (high)gpt-oss-120b (low)Blended price (per 1M tokens)
Total 6 models
Groq offers 6 models, each with different intelligence, performance, and pricing characteristics. Below is a comparison of the key metrics across models.
- For output speed, the fastest models are gpt-oss-20b (high) (866 t/s), gpt-oss-20b (low) (865 t/s), and Llama 3.1 8B (646 t/s). Speed varies significantly across models, with a 84% difference between the fastest and slowest.
- For latency, Llama 3.1 8B (0.84s), Llama 3.3 70B (0.86s), and gpt-oss-20b (high) (3.05s) offer the lowest time to first answer token.
- For pricing, Llama 3.1 8B ($0.05), gpt-oss-20b (low) ($0.10), and gpt-oss-20b (high) ($0.10) offer the lowest blended prices per 1M tokens. Prices vary up to 3.7x across models.
- For context window size, gpt-oss-20b (low) (131k), gpt-oss-20b (high) (131k), and Llama 3.1 8B (131k) support the largest context windows on Groq.
Intelligence Evaluations
Artificial Analysis Intelligence Index
Evaluations
Agentic knowledge work, Elo
Knowledge
1 - hallucination rate
Agentic SaaS workflows
Agentic scientific research workflows in a terminal
Agentic software engineering in open-source repositories
Reasoning & knowledge
Professional document reasoning, All-pass
Agentic coding & terminal use
Physics reasoning
Quantitative analysis on spreadsheets & documents
Legal agentic work, Hallucination-Gated All-Pass Rate
Agentic business operations
Agentic real-world work tasks, Elo
Agentic interactive 3D experiences, Elo
Product and brand videos designed in code, Elo
Kubernetes incident root-cause analysis
Medical long context reasoning
Visual reasoning
Rich, varied, well-crafted language
Well-structured, scannable, answer-first
Truthful, acknowledges uncertainty, non-sycophantic
Use of filler and generic language
Intelligence Index vs. Price
Context Window
Context Window
Pricing
Intelligence Index vs. Price
Performance Summary
Output Speed vs. Price
Speed
Measured by Output Speed (tokens per second)
Output Speed
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
Cache Behaviour
Cache Hit Rate
Cost per Task vs. Cache Hit Rate
End-to-End Response Time
Seconds to output 500 tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time vs. Price
Further Analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
gpt-oss-20b (low) | 131k | Open | -- | -- | 865 | 0.77 | 3.66 | 2.31 | |||
gpt-oss-20b (high) | 131k | Open | -- | -- | 866 | 0.74 | 3.63 | 2.31 | |||
Llama 3.1 8B | 131k | Open | -- | -- | 646 | 0.84 | 1.61 | -- | |||
Llama 3.3 70B | 131k | Open | -- | -- | 327 | 0.86 | 2.39 | -- | |||
gpt-oss-120b (low) | 131k | Open | -- | -- | 471 | 0.67 | 5.98 | 4.25 | |||
gpt-oss-120b (high) | 131k | Open | -- | -- | 471 | 0.63 | 5.94 | 4.25 | |||
Key definitions
Frequently Asked Questions
Common questions about Groq
Groq offers 6 models that we track: gpt-oss-20b (low), gpt-oss-20b (high), Llama 3.1 8B, Llama 3.3 70B, gpt-oss-120b (low), and gpt-oss-120b (high).
The fastest model on Groq by output speed is gpt-oss-20b (high) at 866.2 tokens per second.
The model with the lowest time to first answer token on Groq is Llama 3.1 8B at 0.84s. Lower latency means faster initial response time.
The most affordable model on Groq by blended price is Llama 3.1 8B at $0.05 per 1M tokens (7:2:1 cache hit/input/output ratio).
Prices on Groq vary up to 12x across models, from $0.05 per 1M tokens for Llama 3.1 8B to $0.61 per 1M tokens for Llama 3.3 70B.
Yes, Groq offers an OpenAI-compatible API, making it easy to switch from OpenAI or use existing OpenAI SDK integrations.
Yes, all 6 models on Groq support JSON mode for structured output.
Yes, all 6 models on Groq support function calling (tool use).
Yes, Groq offers 4 reasoning models: gpt-oss-20b (low), gpt-oss-20b (high), gpt-oss-120b (low), and gpt-oss-120b (high). Reasoning models use extended thinking to work through complex problems before providing an answer.
Yes, all 6 models on Groq are open weight models.
Yes, provider performance can vary over time due to infrastructure changes, load balancing, and updates. We continuously benchmark all providers and display historical performance trends in the "Over Time" charts.
When choosing a model on Groq, consider: intelligence (for quality-sensitive tasks), output speed (for throughput-intensive tasks), latency (for interactive applications requiring quick first responses), pricing (for cost-sensitive workloads), and features like context window size, JSON mode, or function calling support.