Together AI

Llama 3.1 405B (Together)

Complete specs, pricing, and benchmark data for Llama 3.1 405B (Together) by Together AI. Last verified 2026-04-20.

JSON ModeFunctionsStreaming
Pricing

Input / 1M tokens

$3.50

Output / 1M tokens

$3.50

Context & Output

Context Window

4K

Max Output

4,096

TTFT

150ms

Speed

100 tok/s

Benchmarks

Arena ELO

1240

Coding ELO

1200

Reasoning ELO

1250

HumanEval

89.5

MMLU

85.9

MATH

112

GPQA

45

Price History (Input $/M tokens)

Strengths
  • +Largest model
  • +High quality
  • +Good pricing
Limitations
  • -Small context window
  • -Slow inference

Best For

General PurposeReasoning

Compare Llama 3.1 405B (Together) with...

Official Pricing Page →
Your ad here