Token cost
USD per 1M tokens, 50% input + 50% output. Lower is better.
LithosAI models
0 models in this catalog are created by LithosAI; 3 additional models are available through its hosting offerings. The newest dated release listed is DeepSeek V4.1 Flash 1017 (Oct 17, 2026).
Updated: Data as of:
Compare hosting providers
DeepSeek V4.1 Flash 1017(max)
Output speed
Tokens per second. Higher is better.
Time to first token
Seconds until the first token. Lower is better.
All models
3 models| Model | Date released | Model details |
|---|---|---|
| DeepSeek V4.1 Flash 1017 | View model | |
| GLM-5.3 | View model | |
| Kimi K3 | View model |