MMeta · Released Dec 6, 2024
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...
- Released
- Dec 6 2024
- Context window
- 131K 16K out
- Input / 1M tokens
- $0.10 from Free
- Output / 1M tokens
- $0.32 from Free
Where to run it
48 providers compared
Free to try on Pendra, Llama — free endpoints are often limited-time and rate-limited. Cheapest: Meganova at $0.10 / $0.30 per 1M input / output tokens — 95% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| Pendrafree · may be limited-time | Free | Free | — | 128K | 4K | — | Free |
| Llamafree · may be limited-time | Free | Free | — | 128K | 4K | — | Free |
| Meganova | $0.10 | $0.30 | — | 131K | 16K | — | Cheapest |
| DeepInfraFP8 | $0.10 | $0.32 | — | 131K | 16K | 99.7%24h | +3% |
| Kilo Gateway | $0.10 | $0.32 | — | 131K | 16K | — | +3% |
| NovitaBF16 | $0.14 | $0.40 | — | 12K | 11K | 99.9%24h | +34% |
| LLM Gateway | $0.14 | $0.40 | — | 131K | 120K | — | +34% |
| Novita AI | $0.14 | $0.40 | — | 12K | 12K | — | +34% |
| DevPass (LLM Gateway) | $0.14 | $0.40 | — | 131K | 4K | — | +34% |
| AkashMLFP8 | $0.20 | $0.52 | $0.10 | 131K | 128K | 100%24h | +87% |
| ParasailFP8 · no tool calling | $0.22 | $0.50 | $0.11 | 131K | 16K | 99.9%24h | +93% |
| Merge Gateway | $0.22 | $0.50 | $0.11 | 131K | 33K | — | +93% |
| Crusoe | $0.25 | $0.75 | $0.13 | 128K | 4K | — | +150% |
| SambaNovano tool calling | $0.45 | $0.90 | — | 131K | 3K | 99.9%24h | +275% |
| STACKIT | $0.53 | $0.76 | — | 128K | 4K | — | +292% |
| Groq | $0.59 | $0.79 | $0.29 | 131K | 33K | 100%24h | +327% |
| Helicone | $0.59 | $0.79 | — | 131K | 33K | — | +327% |
| Abacus | $0.59 | $0.79 | — | 128K | 33K | — | +327% |
| Hugging Face | $0.59 | $0.79 | — | 131K | 4K | — | +327% |
| DigitalOcean | $0.65 | $0.65 | — | 128K | 4K | — | +333% |
| IO.NET | $0.55 | $1.02 | $0.27 | 128K | 4K | — | +344% |
| CoreWeaveFP16 | $0.71 | $0.71 | $0.71 | 128K | 115K | 100%24h | +373% |
| Azure | $0.71 | $0.71 | — | 128K | 33K | — | +373% |
| Azure Cognitive Services | $0.71 | $0.71 | — | 128K | 33K | — | +373% |
| Googleus-central1 | $0.72 | $0.72 | — | 128K | 8K | — | +380% |
| Googleno tool calling | $0.72 | $0.72 | — | 128K | 115K | — | +380% |
| Vercel AI Gateway | $0.72 | $0.72 | — | 128K | 8K | — | +380% |
| Amazon Bedrock | $0.72 | $0.72 | — | 128K | 4K | — | +380% |
| OCI Generative AIno tool calling | $0.72 | $0.72 | — | 128K | 4K | — | +380% |
| Cortecs | $0.72 | $0.72 | — | 131K | 131K | — | +383% |
| OVHcloud AI Endpoints | $0.74 | $0.74 | — | 131K | 131K | — | +393% |
| Neon | $0.50 | $1.50 | — | 128K | 8K | — | +400% |
| watsonx.ai | $0.75 | $0.75 | — | 131K | 4K | — | +402% |
| CloudflareFP8 · no tool calling | $0.29 | $2.25 | — | 24K | 22K | 100%24h | +422% |
| Cloudflare Workers AI | $0.29 | $2.25 | — | 24K | 24K | — | +422% |
| Pioneer | $0.90 | $0.90 | $0.90 | 16K | 16K | — | +500% |
| Scaleway | $1.01 | $1.01 | — | 100K | 16K | — | +573% |
| Eden AI | $1.01 | $1.01 | — | 128K | 4K | — | +574% |
| Together | $1.04 | $1.04 | — | 131K | 2K | 98.6%24h | +593% |
| Together AI | $1.04 | $1.04 | — | 131K | 131K | — | +593% |
| Regolo AI | $0.60 | $2.70 | — | 128K | 16K | — | +650% |
| evroc | $1.15 | $1.15 | — | 128K | 4K | — | +667% |
| Venice AI | $0.70 | $2.80 | — | 128K | 4K | — | +717% |
| GreenPT | $1.25 | $1.25 | — | 100K | 16K | — | +736% |
| Prime Inference | $1.04 | $2.25 | — | 131K | 16K | — | +795% |
| AI/ML API | $1.35 | $1.35 | — | 128K | 127K | — | +801% |
| NanoGPTno tool calling | $1.75 | $2.75 | $1.75 | 128K | 16K | — | +1233% |
| Tinfoil | $1.75 | $2.75 | — | 131K | 4K | — | +1233% |
| CloudFerro Sherlock | $2.92 | $2.92 | — | 70K | 70K | — | +1847% |
Value
Price vs. performance
Llama 3.3 70B Instruct Other Meta Other labs Value frontier · GPQA Diamond vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
7 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 26.0% | #98of 102 | Claude Opus 5.566.9% |
| Fiction.LiveBenchWriting & content · Understanding long creative writing (16k tokens) | 33.3% | #22of 24 | GPT-597.2% |
| GPQA DiamondReasoning & knowledge · PhD-level biology, chemistry and physics questions | 47.4% | #108of 122 | Gemini 4 Argon96.4% |
| SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people | 19.9% | #52of 58 | Claude Fable 581.9% |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 0.0% | #92of 111 | GPT-5.6 Sol32.3% |
| LMCAReasoning & knowledge · Judge the quality of conceptual arguments like experts | 17.5% | #98of 123 | Claude Sonnet 5.568.3% |
| OTIS Mock AIMEMaths · Competition-style problems harder than MATH Level 5 | 5.1% | #113of 120 | Gemini 4 Argon100.0% |
Specs
Details
- Model ID
meta-llama/llama-3.3-70b-instruct- Developer
- Meta
- Release date
- Dec 6, 2024 ·
- Context window
- 131,072 tokens
- Max output
- 16,384 tokens
- API price
- $0.10 / $0.32 per 1M input / output tokens
- Providers
- 48
- Input types
- text
- Reasoning
- No
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.