Llama 3.3 70B (FP8) pricing
Llama 3.3 70B (FP8), Mid-tier from Parasail, costs $0.22 per 1M input tokens and $0.50 per 1M output tokens.
Input$0.22per 1M tokens
Output$0.50per 1M tokens
Context window—tokens
TierMidParasail
Price your own workload against Llama 3.3 70B (FP8) in the calculator →
Specs and pricing
| Provider | Parasail |
| Tier | Mid |
| Context window | — |
| Max output | — |
| Input price | $0.22 / 1M tokens |
| Output price | $0.50 / 1M tokens |
| Cache read | $0.11 / 1M tokens |
| Modalities | text |
Note. Parasail elastic-endpoint serverless pricing, listed by the seller as price per 1M tokens: $0.22/1M input, $0.5/1M output, $0.11/1M cache read; no conversion needed.
What it costs at realistic volumes
Assuming a 4:1 input:output token ratio, a common shape for chat and RAG-style workloads:
| Volume (4:1 in:out ratio) | Input cost | Output cost | Total / month |
|---|---|---|---|
| 1,000,000 in / 250,000 out | $0.22 | $0.13 | $0.34 |
| 10,000,000 in / 2,500,000 out | $2.20 | $1.25 | $3.45 |
| 100,000,000 in / 25,000,000 out | $22.00 | $12.50 | $34.50 |
Cheaper alternatives
Same tier (Mid), a different provider, and a lower blended cost at the 4:1 ratio used above:
| Model | Provider | $/1M in | $/1M out | At the 4:1 ratio |
|---|---|---|---|---|
| GPT-OSS 120B | OpenRouter | $0.037 | $0.17 | 77% cheaper blended |
| GPT-OSS 120B | Novita AI | $0.05 | $0.25 | 67.4% cheaper blended |
| Llama-3.3-70B-Instruct-Turbo | DeepInfra | $0.10 | $0.32 | 47.8% cheaper blended |
| Nemotron 3 Super | OpenRouter | $0.085 | $0.40 | 46.4% cheaper blended |
| DeepSeek V4 Flash 0731 | Baseten | $0.13 | $0.26 | 43.5% cheaper blended |