Deepseek R1 Distill Llama 70B
We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With RL, DeepSeek-R1-Zero naturally emerged with numerous powerful and interesting reasoning behaviors. However,…
76 subscribers
147 ms avg latency
1 endpoints
The in-depth APIMemo review for this API hasn't been published yet —
the data below comes straight from the public marketplace listing.
Deepseek R1 Distill Llama 70B endpoints
| Method | Endpoint | Description |
|---|---|---|
| chat_completions | ||
| POST |
Chat Completions /chat_completions |
Chat Completions |
Deepseek R1 Distill Llama 70B pricing
| Plan | Price | Rate limit | Quotas |
|---|---|---|---|
| BASIC | Free | — |
|
| PRO | $10 / month | 5 / second |
|
| ULTRA | $30 / month | 5 / second |
|
| MEGA | $200 / month | 10 / second |
|
