Groq
Groq is an AI model provider.Tokenando tracks 11 Groq models, with input pricing from $0.050/M and an average blended cost of $0.676/M. Its flagship model is Qwen2.5-72B (Groq).
Custom LPU silicon delivering 5-10x the throughput of GPU inference. Hosts open-weights models at industry-leading speeds.
MODELS TRACKED
11
3 categories
FLAGSHIP
Qwen2.5-72B (Groq)
Live API
MIN INPUT
$0.050/M
cheapest model in family
AVG BLENDED
$0.676/M
across 11 priced models
MAX CONTEXT
128,000
largest window in family
Frontier
1 model
Reasoning
1 model
Efficient
9 models
Llama 3.3 70B (Groq)profile
Live API · 128,000 ctx
in $0.590/Mout $0.790/M
Fastest 70B · manual-seed
Llama 3.1 70B (Groq)profile
Live API · 128,000 ctx
in $0.590/Mout $0.790/M
70B on LPU · manual-seed
Llama 3.1 8B (Groq)profile
Live API · 128,000 ctx
in $0.050/Mout $0.080/M
Fastest 8B · manual-seed
Llama 3.2 11B Vision (Groq)profile
Live API · 128,000 ctx
in $0.180/Mout $0.180/M
Vision on LPU · manual-seed
Llama 3.2 3B (Groq)profile
Live API · 128,000 ctx
in $0.060/Mout $0.060/M
Micro on LPU · manual-seed
Mixtral 8x7B (Groq)profile
Live API · 32,000 ctx
in $0.240/Mout $0.240/M
MoE on LPU · manual-seed
Gemma 2 9B (Groq)profile
Live API · 8,000 ctx
in $0.200/Mout $0.200/M
Google Gemma fast · manual-seed
Qwen2.5-72B (Groq)profile
Live API · 128,000 ctx
in $0.790/Mout $0.790/M
Qwen on LPU · manual-seed
Qwen2.5-Coder-32B (Groq)profile
Live API · 128,000 ctx
in $0.790/Mout $0.790/M
Code on LPU · manual-seed
Frequently Asked Questions
How many models does Groq offer?
Tokenando tracks 11 Groq models.
How much do Groq models cost?
Groq model input pricing starts at $0.050 per million tokens, with an average blended cost of $0.676 per million across the 11 priced models we track.
What is Groq's flagship model?
Groq's flagship model is Qwen2.5-72B (Groq). It is the highest-tier Groq model we track, with input pricing of $0.790 per million tokens.
What model categories does Groq cover?
Groq covers 3 categories: frontier, reasoning and efficient.