Groq

Llama 3.1 8B (Groq)

Complete specs, pricing, and benchmark data for Llama 3.1 8B (Groq) by Groq. Last verified 2026-04-20.

JSON ModeFunctionsStreaming
Pricing

Input / 1M tokens

$0.050

Output / 1M tokens

$0.080

Context & Output

Context Window

128K

Max Output

4,096

TTFT

150ms

Speed

100 tok/s

Benchmarks

Arena ELO

1120

Reasoning ELO

1120

HumanEval

72

MMLU

79

MATH

40

GPQA

40

Price History (Input $/M tokens)

Strengths
  • +Insanely fast
  • +Super cheap
  • +Best for real-time
Limitations
  • -Small model
  • -Limited reasoning

Best For

Fast InferenceCost Effective

Compare Llama 3.1 8B (Groq) with...

Official Pricing Page →
Your ad here