Side-by-side technical and economic comparison between Meta / Together's Llama 3.1 405B — Together AI and OpenAI's OpenAI o1.
• Llama 3.1 405B — Together AI: Classified as a Legacy Reference model (Deprecated on Together AI (Reference)).
• OpenAI o1: Classified as a Legacy Reference model (Generally Available).
| Metric | Llama 3.1 405B — Together AI | OpenAI o1 | Advantage |
|---|---|---|---|
| Provider Organization | Meta / Together | OpenAI | — |
| Standard Input / 1M | $3.50 | $15.00 | Llama 3.1 405B — Together AI (77% lower) |
| Cached Input / 1M | None | $7.50 | Llama 3.1 405B — Together AI lower |
| Output / 1M | $3.50 | $60.00 | Llama 3.1 405B — Together AI lower |
| Context Window | 128k | 200k | OpenAI o1 (200k) |
| Max Generation Tokens | 4.1k | 100k | OpenAI o1 |
| Tokenizer Family | Meta Llama 3 Tiktoken BPE (~128k vocabulary) | OpenAI o200k_base (200k vocabulary) | — |
Opt for Llama 3.1 405B — Together AI when your engineering requirements prioritize Meta / Together's ecosystem, specific tokenizer efficiencies (Exact Llama 3 Tokenizer (±2%)), or when your expected prompt-to-completion ratios favor its $3.5/M input rate.
Opt for OpenAI o1 when looking for OpenAI's tooling integration, specific context window depth (200k tokens), or when output generation volume favors its $60/M rate.