Llama 3.3 70B Instruct pricing & price history
Llama 3.3 70B Instruct by Meta Llama costs $0.71 per 1M input tokens and $0.71 per 1M output tokens, with a 131K token context window.
Price history
| Date | Input | Output |
|---|---|---|
| 2026-11-10 | $0.59 → $0.71 +20% | $0.79 → $0.71 -10% |
| 2026-11-06 | $0.10 → $0.59 +490% | $0.32 → $0.79 +147% |
| 2026-11-04 | $0.20 → $0.10 -50% | $0.52 → $0.32 -38% |
| 2026-11-03 | $0.10 → $0.20 +100% | $0.32 → $0.52 +63% |
| 2026-10-16 | $0.59 → $0.10 -83% | $0.79 → $0.32 -59% |
| 2026-10-13 | $0.22 → $0.59 +168% | $0.50 → $0.79 +58% |
| 2026-10-09 | $0.10 → $0.22 +120% | $0.32 → $0.50 +56% |
| 2026-10-08 | $0.22 → $0.10 -55% | $0.50 → $0.32 -36% |
| 2026-10-05 | $0.10 → $0.22 +120% | $0.32 → $0.50 +56% |
| 2026-10-05 | $0.22 → $0.10 -55% | $0.50 → $0.32 -36% |
| 2026-10-04 | $0.10 → $0.22 +120% | $0.32 → $0.50 +56% |
Cost calculator
Cheaper alternatives to Llama 3.3 70B Instruct
| Model | Input | Output | Context | Intel. | |
|---|---|---|---|---|---|
 GLM 5.3 Flash Z.ai (Zhipu) | $0.0399 | $0.0784 | 1M | 27.8 | compare |
 DeepSeek V4 Flash 0731 DeepSeek | $0.0434 | $0.30 | 1M | — | compare |
DeepSeek V4.1 Flash DeepSeek | $0.022 | $0.40 | 1M | — | compare |
DeepSeek V4.1 Flash 1017 DeepSeek | $0.098 | $0.29 | 1M | 28.6 | compare |
Llama 4 Scout Meta Llama | $0.10 | $0.30 | 1.3M | — | compare |
 MiMo-V2.6-Flash Xiaomi | $0.14 | $0.28 | 1.1M | 26.7 | compare |
Llama Guard 4 12B Meta Llama | $0.18 | $0.18 | 164K | — | compare |
 Claude Haiku 5.5 Anthropic | $0.10 | $0.50 | 1M | 36.5 | compare |
Using a relay or third-party API?
Check for free whether it really serves Llama 3.3 70B Instruct: output fingerprint, tokenizer and billing evidence, results in 1–3 minutes.
About Llama 3.3 70B Instruct
The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...
FAQ
How much does Llama 3.3 70B Instruct cost?
Llama 3.3 70B Instruct by Meta Llama costs $0.71 per 1M input tokens and $0.71 per 1M output tokens, with a 131K token context window. A typical request with 2,000 input and 500 output tokens costs about $0.0018.
What is the context window of Llama 3.3 70B Instruct?
Llama 3.3 70B Instruct supports up to 131,072 tokens of context and up to 115,200 output tokens per request.
Has Llama 3.3 70B Instruct’s price changed?
Yes. ModelBank recorded 11 price changes; the latest on 2026-11-10.
Is Llama 3.3 70B Instruct being deprecated?
No retirement date has been announced for Llama 3.3 70B Instruct as of 2026-11-12.
What are cheaper alternatives to Llama 3.3 70B Instruct?
Cheaper options include GLM 5.3 Flash ($0.0399/$0.0784), DeepSeek V4 Flash 0731 ($0.0434/$0.30), DeepSeek V4.1 Flash ($0.022/$0.40), DeepSeek V4.1 Flash 1017 ($0.098/$0.29).