Groq vs SambaNova Cloud pricing
Monthly cost at three volumes
| Volume | Groq | SambaNova Cloud |
|---|---|---|
| small1M input, 200K output | $0.036 / moLlama Prompt Guard 2 22M | $0.338 / mogpt-oss-120b |
| growing50M input, 10M output | $1.80 / moLlama Prompt Guard 2 22M | $16.90 / mogpt-oss-120b |
| scale1B input, 200M output | $36 / moLlama Prompt Guard 2 22M | $338 / mogpt-oss-120b |
| your volumeset above | $1.80 / moLlama Prompt Guard 2 22M | $16.90 / mogpt-oss-120b |
Each cell is the cheapest feasible plan, or the chosen model, at that volume, from the shared Overage cost function. "Cheapest model" only considers models with both an input and an output price listed. Volumes are example scenarios you can edit.
Cheapest model per service, USD per 1M tokens
| Service | Cheapest model | Input | Output | Models listed |
|---|---|---|---|---|
| Groq | Llama Prompt Guard 2 22M | $0.03 | $0.03 | 6 |
| SambaNova Cloud | gpt-oss-120b | $0.22 | $0.59 | 6 |
Every model
| Model | Service | Input | Output |
|---|---|---|---|
| Llama Prompt Guard 2 22M | Groq | $0.03 | $0.03 |
| Prompt Guard 2 86M | Groq | $0.04 | $0.04 |
| GPT OSS 20B | Groq | $0.075 | $0.30 |
| Safety GPT OSS 20B | Groq | $0.075 | $0.30 |
| GPT OSS 120B | Groq | $0.15 | $0.60 |
| Qwen/Qwen3.8-27B | Groq | $0.80 | $4 |
| gpt-oss-120b | SambaNova Cloud | $0.22 | $0.59 |
| MiniMax-M3.1-Flash | SambaNova Cloud | $0.30 | $1.20 |
| gemma-4-31B-it | SambaNova Cloud | $0.38 | $1.15 |
| Meta-Llama-3.3-70B-Instruct | SambaNova Cloud | $0.60 | $1.20 |
| DeepSeek-V3.1 | SambaNova Cloud | $3 | $4.50 |
| DeepSeek-V3.2 | SambaNova Cloud | $3 | $4.50 |
Models are compared by name and price, never by position. Pick one model per service, or "Cheapest model", in the cost table above.
Tightest rate limit
| Measure | Groq | SambaNova Cloud |
|---|---|---|
| Tightest limit | 100 api requests per minuteLlama Prompt Guard 2 22M tier | unlisted |
Changes in the last 90 days
No changes recorded.
No change detected for either service in the last 90 days.