Compare verified pricing and model limits for a low-cost text model and a multimodal option available through GroqCloud. The cost example uses a structured data extraction workload and does not assume equal model quality.
Llama 4 Scout accepts text and images on the tracked GroqCloud endpoint; GPT-OSS 20B is tracked as text-only. That functional boundary may decide the shortlist before the small rate difference matters.
| Decision factor | GPT-OSS 20B on Groq | Llama 4 Scout on Groq |
|---|---|---|
| Model creator | OpenAI | Meta |
| Accepted input | Text | Text and images |
| Maximum output | 65,536 tokens | 8,192 tokens |
| GroqCloud token price | $0.075 input and $0.30 output per million | $0.11 input and $0.34 output per million |
Prices are USD per 1M tokens under each model's verified default profile.
| Field | GPT-OSS 20B on Groq | Llama 4 Scout on Groq |
|---|---|---|
| API access provider | GroqCloud | GroqCloud |
| Model creator | OpenAI | Meta |
| API model ID | openai/gpt-oss-20b | meta-llama/llama-4-scout-17b-16e-instruct |
| Input / 1M | $0.075 | $0.11 |
| Cached input / 1M | $0.037 | Unavailable |
| Output / 1M | $0.30 | $0.34 |
| Context window | 131,072 tokens | 131,072 tokens |
| Maximum output | 65,536 tokens | 8,192 tokens |
| Accepted input | text | text, image |
4,000 input and 500 output tokens per request, 100,000 monthly requests, and 25% cached input.
Cost result: GPT-OSS 20B on Groq is $19.80 lower per month for these assumptions. This is a price comparison, not a model-quality ranking.
Document parsing, OCR, validation infrastructure, and failed-record handling are outside this token estimate.
Latency, reliability, output quality, retries, regional processing, and provider-specific tool charges can change the practical decision.
These comparisons share a model or provider decision with this page.