Llama 3.1 405 B Instruct pricing
Source: litellm · updated 2026-11-17
Short answer: Llama 3.1 405 B Instruct costs $3.50 per million input tokens and $3.50 per million output tokens. That is a 3:1 blended rate of $3.50 per million tokens, ranking it #747 of 823 tracked models by input price.
Input / 1M
$3.50
Output / 1M
$3.50
Output ÷ input
1.00×
Context window
4,096
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $3.50 |
| Output, per 1M tokens | $3.50 |
| Output ÷ input multiple | 1.00× |
| 3:1 blended, per 1M tokens | $3.50 |
| Context window | 4,096 tokens |
| Vendor | together_ai |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total |
|---|---|---|---|
| 1M in / 250K out | $3.50 | $0.875 | $4.38 |
| 10M in / 2.5M out | $35.00 | $8.75 | $43.75 |
| 100M in / 25M out | $350 | $87.50 | $438 |
Other together_ai models
| Model | Input | Output |
|---|---|---|
| Qwen 2 1.5 B Instruct | $0.020 | $0.020 |
| Trinity Mini | $0.045 | $0.150 |
| Nvidia Nemotron Nano 9 B | $0.060 | $0.250 |
| Qwen 2 1.5 B Instruct | $0.100 | $0.100 |
| Nvidia Nemotron 3.5 Super VL | $0.150 | $0.500 |
Cheaper alternatives
Next cheaper option: Gemini 2.5 Pro at $1.25 per 1M input tokens. Next step up: Pareto 26.9 at $2.50 per 1M input tokens.
Spotted a price that is out of date? Submit a correction.