Nova Lite (APAC) vs Meta: Llama 3.3 70B Instruct
Meta: Llama 3.3 70B Instruct is cheaper than Nova Lite (APAC) by about 100% on a blended (75% input / 25% output) basis. Nova Lite (APAC) charges $0.060000 / $0.240000 per 1M input/output tokens; Meta: Llama 3.3 70B Instruct charges $0.000000 / $0.000000.
| Field | Nova Lite (APAC) | Meta: Llama 3.3 70B Instruct |
|---|---|---|
| Input $/Mtok | $0.060000 | $0.000000 |
| Output $/Mtok | $0.240000 | $0.000000 |
| Cache read $/Mtok | – | – |
| Cache write $/Mtok | – | – |
| Input $/Mchar | n/a | 0.000000 |
| Output $/Mchar | n/a | 0.000000 |
| Context window | 300000 | 128000 |
| Tokenizer | – | llama3 |
Price history
Same workload, monthly cost
Illustrative monthly cost at 10M tokens/month, split per workload preset. Not a substitute for the calculator, which uses your own volume, cache and batch assumptions.
| Workload | Nova Lite (APAC) | Meta: Llama 3.3 70B Instruct |
|---|---|---|
| Coding agent | $0.69 | $0.00 |
| RAG chat | $0.87 | $0.00 |
| Summarisation (batch) | $0.78 | $0.00 |
| Classification | $0.64 | $0.00 |
| Chat assistant | $1.32 | $0.00 |
| Content generation | $2.04 | $0.00 |
Cite this page
Nova Lite (APAC) vs Meta: Llama 3.3 70B Instruct: $0.060000/$0.240000 vs $0.000000/$0.000000 per 1M tokens, as of 2026-10-04 (Token Barometer, https://tokenbarometer.com/compare/amazon_nova-lite-vs-meta_llama-3.3-70b-instruct).
Nova Lite (APAC) page | Meta: Llama 3.3 70B Instruct page | Open in calculator