Flowers Labs
Upload your input data and generate synthetic data for cheap. Tokens wasted due to RLHF rejections are not counted.
Model | Endpoint | Input (USD) | Output (USD) |
---|
Mistral 7B | mistral-7b | 0$ / 1M tokens | 0.1$ / 1M tokens |
Mistral 7B Instruct | mistral-7b-i | 0$ / 1M tokens | 0.1$ / 1M tokens |
Mixtral 8x7B | mixtral-8x7b | 0$ / 1M tokens | 0.25$ / 1M tokens |
Mixtral 8x7B Instruct | mixtral-8x7b-i | 0$ / 1M tokens | 0.25$ / 1M tokens |
Flowers Small | flowers-small | 3$ / 1M tokens | 10$ / 1M tokens |
Flowers Medium | flowers-medium | 10$ / 1M tokens | 45$ / 1M tokens |
- 1320 FP16 TFLOPS
- CUDA support (Nvidia GPUs yay!)
- 96 GB 1TB/s GPU memory
- 28 GB/s NVMe read bandwidth
- 8 TB NVMe and 256 GB RAM
- AMD Epyc 32 cores
- 2x 25 GbE + 2x 10 GbE
- 10 SATA 6 Gb/s ports
- $12,000 (shipping Q3 2024)