SambaNova models
SambaNova Cloud inference with high-throughput open models and public per-token rates.
GPT OSS 120B (SambaNova)
$0.22 / $0.59 per 1M tokens
Gemma 4 31B (SambaNova)
$0.38 / $1.15 per 1M tokens
Llama 3.3 70B (SambaNova)
$0.6 / $1.20 per 1M tokens
MiniMax M2.7 (SambaNova)
$0.6 / $2.40 per 1M tokens
SambaNova
SambaNova Cloud inference with high-throughput open models and public per-token rates.
There are 4 published models here. On a 70/30 mix, combined list rates run from $0.331 to $1.14 per 1 million tokens, with a median of $0.6955. The largest listed context window in this group is 200,000 tokens. The most recent price check in this group is 2026-08-14.
Official SambaNova pricing page
SambaNova list rates
USD per million tokens. The 70/30 mix is a planning default — change it on the model page if your traffic is output-heavy.
| Model | Input / 1M | Output / 1M | Mixed 70/30 | Context | Verified |
|---|---|---|---|---|---|
| GPT OSS 120B (SambaNova) | $0.22 | $0.59 | $0.331 | 131,072 | |
| Gemma 4 31B (SambaNova) | $0.38 | $1.15 | $0.611 | 128,000 | |
| Llama 3.3 70B (SambaNova) | $0.6 | $1.20 | $0.78 | 128,000 | |
| MiniMax M2.7 (SambaNova) | $0.6 | $2.40 | $1.14 | 200,000 |
Questions about SambaNova rates
Do these SambaNova figures include discounts and tax?
No. They are public list rates. Commits, credits, regions, tax, and commercial discounts are not in the number — check the source linked on each model.
When is the more expensive SambaNova tier worth the extra?
MiniMax M2.7 (SambaNova) sits near $1.14 per 1M tokens on a 70/30 mix, versus $0.331 on GPT OSS 120B (SambaNova). Use the high tier when retries or long reasoning actually fail on the cheap one.