Difference between revisions of "AI compute"
KevinYager (talk | contribs) (→Cloud LLM Routers) |
KevinYager (talk | contribs) (→Cloud LLM Routers) |
||
| Line 11: | Line 11: | ||
* [https://glaive.ai/ Glaive AI] | * [https://glaive.ai/ Glaive AI] | ||
| − | ==Cloud LLM Routers== | + | ==Cloud LLM Routers & Inference Providers== |
* [https://openrouter.ai/ OpenRouter] | * [https://openrouter.ai/ OpenRouter] | ||
* [https://www.litellm.ai/ LiteLLM] | * [https://www.litellm.ai/ LiteLLM] | ||
* [https://centml.ai/ Cent ML] | * [https://centml.ai/ Cent ML] | ||
* [https://fireworks.ai/ Fireworks AI] | * [https://fireworks.ai/ Fireworks AI] | ||
| + | * Huggingface [https://huggingface.co/blog/inference-providers Inference Providers Hub] | ||
==Acceleration Hardware== | ==Acceleration Hardware== | ||
Revision as of 09:51, 29 January 2025
Contents
Cloud GPU
Cloud Training Compute
Cloud LLM Routers & Inference Providers
- OpenRouter
- LiteLLM
- Cent ML
- Fireworks AI
- Huggingface Inference Providers Hub
Acceleration Hardware
- Nvidia GPUs
- Google TPU
- Etched: Transformer ASICs
- Cerebras
- Untether AI
- Graphcore
- SambaNova Systems
- Groq
- Tesla Dojo
- Deep Silicon: Combined hardware/software solution for accelerated AI (e.g. ternary math)
