Fireworks AI

Hosted inference and fine-tuning for open-weights models.

Hosted inference and fine-tuning across hundreds of open-weights models (Llama, Mistral, DeepSeek, Qwen, etc.). Strong pricing for inference-at-scale; LoRA + full fine-tuning supported.

Inference cloud for serving any AI model in production.

Production inference platform offering both pre-optimized Model APIs (Llama, DeepSeek, and more, billed per token) and dedicated GPU/CPU deployments for custom models, billed per minute with no charge for idle time. Custom models are packaged with its open-source Truss format and autoscale, including scale-to-zero. Aimed at low-latency, high-throughput serving.

Low-latency inference for open-weights models on custom LPU chips.

GroqCloud serves open-weights models (Llama, DeepSeek, Qwen, Kimi) on Groq's purpose-built LPU hardware, hitting hundreds of tokens per second where GPUs manage tens. OpenAI-compatible API with a free tier; the default when token latency is the product.

Serverless GPUs. Run training, inference, batch jobs from Python.

Define cloud workloads in Python, deploy with one command — GPU access on demand, fast cold starts, fair-share pricing. The default 'I need to fine-tune a model from a Jupyter cell' platform.