SambaNova Cloud review

SambaNova Cloud is a smaller alternative to Groq and Cerebras that also runs on custom silicon, SambaNova's RDU dataflow chips. The public pricing page listed five models when we checked, with gpt-oss-120b at $0.22 per 1M input tokens and $0.59 per 1M output tokens, MiniMax-M3.1-Flash at $0.30 and $1.20, and GLM-5.3 at $2.80 and $8.80, which is above GPU-based rivals for that model. The API is OpenAI-compatible and documents function calling, JSON mode, vision, embeddings and prompt caching, and custom checkpoints let you serve your own fine-tuned weights. Plans are simple: the Free plan needs a payment method and purchased credits, the Developer plan is pay as you go with standard rate limits, and Enterprise is a subscription with production limits and BYOC. The short catalog keeps it in the middle of the list.

Who it is for

Teams that want fast-hardware inference on a short model list

Price

Pay as you go per token after buying credits; gpt-oss-120b at $0.22 per 1M input tokens and $0.59 per 1M output tokens.

gpt-oss-120b N/A 0.22 USD 0.59 USD
Quoted from the vendor pricing page on November 12, 2026.

Key features

• Runs on SambaNova RDU hardware rather than GPUs
• OpenAI-compatible API with function calling, JSON mode, vision, embeddings and prompt caching
• Five priced models when we checked, including GLM-5.3, MiniMax-M3.1-Flash and gpt-oss-120b
• Developer plan with standard rate limits and access to production and preview models
• Enterprise plan with production rate limits, BYOC and custom rate limits

Strengths and weaknesses

• Strength: gpt-oss-120b output at $0.59 per 1M tokens, slightly below Together AI and Fireworks AI
• Strength: Cached input pricing on GLM-5.3 and MiniMax-M3.1-Flash
• Strength: Custom checkpoints let you run your own fine-tuned weights
• Weakness: Short public model list compared with Together AI or DeepInfra
• Weakness: You must buy credits before the first request, even on the Free plan
• Weakness: Higher rate limits and BYOC require the subscription-based Enterprise plan

Key facts

• gpt-oss-120b: $0.22 input, $0.59 output per 1M tokens
• MiniMax-M3.1-Flash: $0.30 input, $1.20 output per 1M tokens
• Plans: Free (credits), Developer (pay as you go), Enterprise (subscription)
• Hardware: SambaNova RDU

Where it ranks

SambaNova Cloud is No. 7 in our Best AI Inference Platforms verdict.