SambaNova

Custom AI chip inference platform with purpose-built hardware for high-throughput LLM serving

Free Trial

SambaNova builds custom AI inference chips (SN40L and SN50) and offers SambaNova Cloud, an inference API for running large language models. Their purpose-built chips deliver significantly higher throughput than NVIDIA GPUs, claiming 5.7x faster than H200 on DeepSeek R1. The cloud API is OpenAI-compatible and supports Llama 3.1 (8B, 70B, 405B) and DeepSeek R1. Free tier includes $5 in credits. Developer and enterprise tiers available with pay-as-you-go token-based pricing.

Pricing: Per token usage

Visit website →

Pricing

Posts

Suggest changes

SambaNova Alternatives

Explore 31 products in the Inference APIs category. View all SambaNova alternatives.

Beam

Open-source serverless GPU cloud with sub-second cold starts and auto-scaling

Vast.ai

GPU marketplace for renting compute at market-driven prices with per-second billing

Predibase

Serverless fine-tuned LLM serving with LoRAX, serving thousands of adapters on a single GPU

CoreWeave

GPU cloud infrastructure built for large-scale AI training and inference workloads

novita.ai

APIs, Serverless and GPU Instance In One AI Cloud

deepinfra

Run the top AI models using a simple API, pay per use. Low cost, scalable and production ready infrastructure.

View all Inference APIs tools ≫

Is your product missing? 👀 Add it here →