Llama 3.3 70 B pricing
Source: litellm · updated 2026-10-30
Short answer: Llama 3.3 70 B costs $0.135 per million input tokens and $0.432 per million output tokens. That is a 3:1 blended rate of $0.209 per million tokens, ranking it #179 of 732 tracked models by input price.
Input / 1M
$0.135
Output / 1M
$0.432
Output ÷ input
3.20×
Context window
128,000
Price breakdown
| Metric | USD |
|---|---|
| Input, per 1M tokens | $0.135 |
| Output, per 1M tokens | $0.432 |
| Output ÷ input multiple | 3.20× |
| 3:1 blended, per 1M tokens | $0.209 |
| Context window | 128,000 tokens |
| Vendor | kyma |
What a typical workload costs
| Monthly volume | Input cost | Output cost | Total |
|---|---|---|---|
| 1M in / 250K out | $0.135 | $0.108 | $0.243 |
| 10M in / 2.5M out | $1.35 | $1.08 | $2.43 |
| 100M in / 25M out | $13.50 | $10.80 | $24.30 |
Other kyma models
| Model | Input | Output |
|---|---|---|
| GPT OSS 120 B | $0.047 | $0.230 |
| Gemma 4 31 B | $0.164 | $0.469 |
| GPT-5.6 Luna Pro | $0.270 | $1.62 |
| Qwen 3 Coder | $0.355 | $1.61 |
| Qwen 3.7 Plus | $0.432 | $1.73 |
Compare against
Cheaper alternatives
Next cheaper option: Morph Dsv 4 Flash at $0.142 per 1M input tokens. Next step up: Lfm 2.5 24 B A2 B at $0.050 per 1M input tokens.
Spotted a price that is out of date? Submit a correction.