Llama 3.3 70B Instruct pricing & benchmarks
Released Dec 6, 2024 · served by 12 providers
What Llama 3.3 70B Instruct is
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...
- Model ID:
meta-llama/llama-3.3-70b-instruct - Modality: text->text
- Knowledge cutoff: 2023-12-31
- Tool calling: supported · Structured output: supported
- Weights:
meta-llama/Llama-3.3-70B-Instructon Hugging Face
Providers and prices for Llama 3.3 70B Instruct
The same model costs different amounts depending on who serves it. Prices are per 1M tokens, cheapest first. Uptime and throughput are OpenRouter's rolling measurements, not vendor claims.
| Provider | Input | Output | Context | Quant | Uptime 24h | Throughput |
|---|---|---|---|---|---|---|
| DeepInfracheapest | $0.100 | $0.320 | 131K | fp8 | 95.4% | — |
| Nebius | $0.130 | $0.400 | 131K | fp8 | 68.3% | — |
| AkashML | $0.130 | $0.400 | 131K | fp8 | 98.2% | — |
| Novita | $0.135 | $0.400 | 6K | bf16 | 89.8% | — |
| Parasail | $0.220 | $0.500 | 131K | fp8 | 97.7% | — |
| CoreWeave | $0.710 | $0.710 | 128K | fp16 | 99.0% | — |
| $0.720 | $0.720 | 128K | — | 100.0% | — | |
| Crusoe | $0.250 | $0.750 | 131K | bf16 | 99.9% | — |
| Groq | $0.590 | $0.790 | 131K | — | 99.9% | — |
| SambaNova | $0.450 | $0.900 | 16K | bf16 | 94.1% | — |
| Together | $1.04 | $1.04 | 131K | fp8 | 97.6% | — |
| Cloudflare | $0.293 | $2.25 | 24K | fp8 | 99.0% | — |
Cheaper models in the same class
Models scoring within 4 points of Llama 3.3 70B Instruct on the Intelligence Index, but with a lower output price.
FAQ
How much does the Llama 3.3 70B Instruct API cost?
$0.100 per 1M input tokens and $0.320 per 1M output tokens. A typical workload of 1M input + 300K output tokens costs about $0.20.
Is Llama 3.3 70B Instruct free?
No. It is a paid model starting at $0.100 per 1M input tokens, though some providers offer trial credits.
Which provider is cheapest for Llama 3.3 70B Instruct?
DeepInfra at $0.100 input / $0.320 output per 1M tokens (fp8 quantization).
Can I self-host Llama 3.3 70B Instruct?
Yes — weights are published as meta-llama/Llama-3.3-70B-Instruct on Hugging Face, so you can run it on your own hardware.