Cost to run
StableLM Zephyr 3B
stabilityai/stablelm-zephyr-3b
- Family
- StableLM
- Context
- 4,096 tokens
StableLM Zephyr 3B can be run self-hosted (rent a GPU + run vLLM/TGI) or through a serverless API (pay per token). Live pricing comparisons:
Self-hosted
Rent a GPU from Lambda, RunPod, Vast.ai, CoreWeave. Full control.
Serverless
Pay per token via Together, Fireworks, DeepInfra, Groq.
Need to benchmark this model against another? Try the calculator or see where it ranks on the InferenceScore leaderboard.