SambaNova LLM API pricing
Every model SambaNova serves, with its pricing per 1M tokens and the exact model ID to copy. Sorted cheapest input first. Verified against LiteLLM and OpenRouter daily.
Models served
19
on this provider
Cheapest input
$0.04
per 1M tokens
Cheapest output
$0.08
per 1M tokens
Max context
10.5M
tokens
| Model ID | |||||
|---|---|---|---|---|---|
| Meta Llama 3.2 1B Instruct | Meta | 16K | $0.04 | $0.08 | sambanova/Meta-Llama-3.2-1B-Instruct |
| Meta Llama 3.2 3B Instruct | Meta | 4K | $0.08 | $0.16 | sambanova/Meta-Llama-3.2-3B-Instruct |
| Meta Llama 3.1 8B Instruct | Meta | 131K | $0.10 | $0.20 | sambanova/Meta-Llama-3.1-8B-Instruct |
| gpt-oss-120b | OpenAI | 131K | $0.22 | $0.59 | sambanova/gpt-oss-120b |
| Meta Llama Guard 3 8B | Meta | 16K | $0.30 | $0.30 | sambanova/Meta-Llama-Guard-3-8B |
| Gemma 4 31B | 262K | $0.38 | $1.15 | sambanova/gemma-4-31B-it | |
| Llama 4 Scout 17B 16e Instruct | Meta | 10.5M | $0.40 | $0.70 | sambanova/Llama-4-Scout-17B-16E-Instruct |
| Qwen3 32B | Alibaba | 131K | $0.40 | $0.80 | sambanova/Qwen3-32B |
| Qwen2 Audio 7B Instruct | Alibaba | 4K | $0.50 | $100.00 | sambanova/Qwen2-Audio-7B-Instruct |
| QwQ 32B | Alibaba | 131K | $0.50 | $1.00 | sambanova/QwQ-32B |
| Meta Llama 3.3 70B Instruct | Meta | 131K | $0.60 | $1.20 | sambanova/Meta-Llama-3.3-70B-Instruct |
| MiniMax M2.7 | MiniMax | 1M | $0.60 | $2.40 | sambanova/MiniMax-M2.7 |
| Llama 4 Maverick 17B 128e Instruct | Meta | 131K | $0.63 | $1.80 | sambanova/Llama-4-Maverick-17B-128E-Instruct |
| R1 Distill Llama 70B | DeepSeek | 131K | $0.70 | $1.40 | sambanova/DeepSeek-R1-Distill-Llama-70B |
| DeepSeek V3 0324 | DeepSeek | 164K | $3.00 | $4.50 | sambanova/DeepSeek-V3-0324 |
| DeepSeek V3.2 | DeepSeek | 164K | $3.00 | $4.50 | sambanova/DeepSeek-V3.2 |
| DeepSeek V3 1 | DeepSeek | 164K | $3.00 | $4.50 | sambanova/DeepSeek-V3.1 |
| R1 | DeepSeek | 164K | $5.00 | $7.00 | sambanova/DeepSeek-R1 |
| Meta Llama 3.1 405B Instruct | Meta | 128K | $5.00 | $10.00 | sambanova/Meta-Llama-3.1-405B-Instruct |
Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI