Model specification · Meta
Llama 3.3 70B Instruct
meta-llama/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...
SpecificationsReported values
- Context window
- 131K
- Maximum output
- 128K
- Architecture
- text->text
- Tokenizer
- Llama3
- Knowledge cutoff
- 2023-12-31
- Moderated
- No
- Input modalities
- text
- Output modalities
- text
Full pricingUSD where reported
| Price | Standard |
|---|---|
| Input | $0.130 / M |
| Output | $0.400 / M |
Benchmark indexesReported indexes
- Intelligence
- 9.4
- Coding
- 11.9
- Agentic
- 0.3
- Design Arena Elo
- —
Example workload costsStandard pricing only
| Workload | Input / output | Cache share | Estimated cost |
|---|---|---|---|
| Quick check | 10K / 1K × 1 | 0% | $0.00170 |
| Long analysis | 100K / 4K × 1 | 50% | $0.015 |
| Workday | 20K / 2K × 1,000 | 30% | $3.40 |
More from MetaAll Meta models
| Model | Context | Input / M | Output / M | Intel. |
|---|---|---|---|---|
| Llama 4 Maverick | 1.0M | $0.200 | $0.800 | 14.3 |
| Llama 4 Scout | 1.3M | $0.100 | $0.300 | 10 |
| Llama 3.1 8B Instruct | 131K | $0.050 | $0.080 | 7.6 |