Llama 4 Scout vs Ministral 3 8B API pricing
Updated
Llama 4 Scout: USD 0.10 input, USD 0.30 output per 1M tokens (DeepInfra). Ministral 3 8B: USD 0.15 input, USD 0.15 output per 1M tokens (Mistral AI and 1 other). At 3 input tokens per output token they cost the same, USD 0.15 per 1M tokens.
How do Llama 4 Scout and Ministral 3 8B differ on price?
- Cheapest input price: Llama 4 Scout USD 0.08 against USD 0.15 per 1M tokens, so Ministral 3 8B costs 1.88× as much (88% more).
- Cheapest output price: Ministral 3 8B USD 0.15 against USD 0.30 per 1M tokens, so Llama 4 Scout costs 2× as much.
- Among the 201 indexed models, Llama 4 Scout ranks 40th cheapest and Ministral 3 8B 40th on the 3:1 blend (the same rank).
- The cheaper model changes at 3 input tokens per output token.
- Only Ministral 3 8B lists a cached-input discount: Ministral 3 8B at 1 standard price of 2, from USD 0.015 (Mistral AI, 90% below that provider's input price). Llama 4 Scout's cached tokens bill at the input price.
- Llama 4 Scout states up to 1,048,576 tokens of context, 4× the 262,144 of Ministral 3 8B.
- Llama 4 Scout has 6 providers with a standard price, 4 more than Ministral 3 8B; 1 provider sells both (Amazon Bedrock).
- On one account, Amazon Bedrock charges USD 0.2925 for Llama 4 Scout (1.95× its cheapest price) and USD 0.15 for Ministral 3 8B (its cheapest price) at 3:1, the lowest combined price among the providers that sell both.
- Shopping around matters more for Llama 4 Scout: its dearest standard price is 2.8× its cheapest, against 1× for Ministral 3 8B.
- Mistral AI's own API is one of 2 providers tied for the lowest standard price we found for Ministral 3 8B, USD 0.15 at 3:1.
- Llama 4 Scout has a batch price at 2 providers, from USD 0.1462 at 3:1 (Amazon Bedrock); Ministral 3 8B has a batch price at 1 provider, from USD 0.075 at 3:1 (Mistral AI).
- For the one-developer coding agent preset, the gap is USD 1,345 a year in Ministral 3 8B's favour (Llama 4 Scout USD 189.00 a month, Ministral 3 8B USD 76.95).
- Per character of input, the gap depends on the language: Llama 4 Scout costs 34% less than Ministral 3 8B on Arabic MSA text but 61% less than Ministral 3 8B on Chinese, because the two tokenizers split those languages differently.
Llama 4 Scout and Ministral 3 8B prices side by side
USD per 1M tokens, standard tier. "Cheapest" is the lowest price at 3 input tokens to 1 output token; each line can come from a different provider.
| Price | Llama 4 Scout | Ministral 3 8B |
|---|---|---|
| Model page | Llama 4 Scout 17B 16E Instruct Meta · 6 providers | Ministral 3 8B Mistral AI · 2 providers |
| Cheapest standard price, blended 3:1 | USD 0.15 tie USD 0.10 in / USD 0.30 out at DeepInfra DeepInfra page checked 2026-11-14 | USD 0.15 tie USD 0.15 in / USD 0.15 out at Mistral AI and 1 other Mistral AI page checked 2026-11-14 |
| Cheapest input | USD 0.08 lower GMI Cloud source | USD 0.15 Mistral AI source |
| Cheapest cached input | No standard price lists a cache price; cached tokens bill at the input price. | USD 0.015 Mistral AI source |
| Cheapest output | USD 0.30 DeepInfra source | USD 0.15 lower Mistral AI source |
| Maker's own API | Meta publishes no paid standard price on its own API that we found. | USD 0.15 / USD 0.15 Mistral AI source |
| Median standard price | USD 0.175 / USD 0.625 input / output, 6 prices | USD 0.15 / USD 0.15 input / output, 2 prices |
| Context window | 1,048,576 tokens, largest stated; from 131,072 at some providers | 262,144 tokens, largest stated |
| Providers with a standard price | 6 | 2 |
- Lower marks the lower price on a line; tie, equal prices. Other tiers and free rows are on the model pages.
Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).
What do Llama 4 Scout and Ministral 3 8B cost for a real workload?
30 days at each model's cheapest standard price for the workload (presets of the cost calculator; cached input at the cached price where listed).
| Workload | Llama 4 Scout | Ministral 3 8B | Difference |
|---|---|---|---|
| Support chatbot 2,000 requests a day, 1,500 input tokens (60% cached), 300 output | USD 14.40 DeepInfra | USD 8.91 Mistral AI | USD 5.49 Ministral 3 8B lower |
| RAG over documents 1,000 requests a day, 8,000 input tokens (30% cached), 500 output | USD 26.70 GMI Cloud | USD 28.53 Mistral AI | USD 1.83 Llama 4 Scout lower |
| Coding agent (one developer) 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning | USD 189.00 GMI Cloud | USD 76.95 Mistral AI | USD 112.05 Ministral 3 8B lower |
The cheaper model depends on the workload: Ministral 3 8B for Support chatbot, Llama 4 Scout for RAG over documents, Ministral 3 8B for Coding agent (one developer). Cache-heavy work rewards a low cached-input price. The widest gap is on Coding agent (one developer): the dearer model costs 2.46× as much as the cheaper one.
Does the cheaper model change with your input:output ratio?
The cheaper model changes with the mix: below 3 input tokens per output token, Ministral 3 8B is cheaper; above 3, Llama 4 Scout is cheaper. At exactly 3:1 they cost the same. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
| Input : output tokens | Llama 4 Scout | Ministral 3 8B |
|---|---|---|
| 1 : 3 (long answers) | USD 0.25 DeepInfra | USD 0.15 lower Mistral AI |
| 1 : 1 | USD 0.20 DeepInfra | USD 0.15 lower Mistral AI |
| 3 : 1 (the site’s blend) | USD 0.15 tie DeepInfra | USD 0.15 tie Mistral AI |
| 10 : 1 (retrieval, long prompts) | USD 0.1182 lower DeepInfra | USD 0.15 Mistral AI |
| 30 : 1 (agents, big contexts) | USD 0.09355 lower GMI Cloud | USD 0.15 Mistral AI |
USD per 1M tokens at the cheapest standard provider for each mix, before cache discounts. Thinking tokens bill as output.
What do Llama 4 Scout and Ministral 3 8B cost per character in other languages?
Cost to send 1M characters at each model's cheapest non-promotional standard input price, with the Llama 4 Maverick tokenizer for Llama 4 Scout and the Ministral 3 14B tokenizer for Ministral 3 8B:
| Content | Llama 4 Scout | Ministral 3 8B |
|---|---|---|
| English | USD 0.017 lower GMI Cloud | USD 0.033 Amazon Bedrock |
| Arabic MSA | USD 0.030 lower GMI Cloud | USD 0.045 Amazon Bedrock |
| Egyptian Arabic | USD 0.031 lower GMI Cloud | USD 0.047 Amazon Bedrock |
| Hindi | USD 0.027 lower GMI Cloud | USD 0.057 Amazon Bedrock |
| Chinese | USD 0.053 lower GMI Cloud | USD 0.135 Amazon Bedrock |
| Spanish | USD 0.018 lower GMI Cloud | USD 0.035 Amazon Bedrock |
| Code | USD 0.020 lower GMI Cloud | USD 0.040 Amazon Bedrock |
From the real cost by language dataset (measured tokens per character on FLORES-200 and other texts). Your own text will differ.
Check it on your own traffic
- Llama 4 Scout 17B 16E Instruct API pricing: all 6 providers and other tiers.
- Ministral 3 8B API pricing: all 2 providers and other tiers.
- Cost calculator with Llama 4 Scout and Ministral 3 8B preloaded
- Llama 4 token counter
- Ministral 3 token counter
- Prompt caching calculator
More comparisons: Llama 4 Scout vs Gemma 4 31B · Ministral 3 14B vs Ministral 3 8B · every Llama 4 Scout comparison · every Ministral 3 8B comparison · all 662 price comparisons
Questions
Is Llama 4 Scout cheaper than Ministral 3 8B?
Not on price: on a blend of 3 input tokens to 1 output token they tie at USD 0.15 per 1M tokens. For input alone, Llama 4 Scout is cheaper (USD 0.08 against USD 0.15 per 1M); for output alone, Ministral 3 8B is (USD 0.15 against USD 0.30). This compares price only, not quality.
What do Llama 4 Scout and Ministral 3 8B cost per 1M tokens?
Llama 4 Scout: from USD 0.10 input and USD 0.30 output at DeepInfra; median of 6 standard prices USD 0.175 / USD 0.625. Ministral 3 8B: from USD 0.15 input and USD 0.15 output at Mistral AI and 1 other; Mistral AI's own API charges USD 0.15 / USD 0.15; median of 2 standard prices USD 0.15 / USD 0.15. Each price is read from the provider's own pricing page; the source is linked in the table.
Which providers sell both Llama 4 Scout and Ministral 3 8B?
1 provider has a standard price for both: Amazon Bedrock (USD 0.2925 and USD 0.15 at 3:1). Buying each model from its own cheapest provider costs USD 0.15 and USD 0.15.
Which is cheaper for a coding agent or a chatbot, Llama 4 Scout or Ministral 3 8B?
For a coding agent (1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning): USD 189.00 a month on Llama 4 Scout (GMI Cloud) against USD 76.95 on Ministral 3 8B (Mistral AI), a difference of USD 112.05. For a support chatbot (2,000 requests a day, 1,500 input tokens (60% cached), 300 output): USD 14.40 a month on Llama 4 Scout (DeepInfra) against USD 8.91 on Ministral 3 8B (Mistral AI), a difference of USD 5.49. Each uses the cheapest standard price for that workload, cached input included. Try your own numbers.
At what input:output ratio does the cheaper of Llama 4 Scout and Ministral 3 8B change?
The cheaper model changes with the mix: below 3 input tokens per output token, Ministral 3 8B is cheaper; above 3, Llama 4 Scout is cheaper. At exactly 3:1 they cost the same. More input per output token, as with retrieval or agents that resend long contexts, favours the lower input price; long answers favour the lower output price.
Do Llama 4 Scout and Ministral 3 8B have cached input pricing?
Llama 4 Scout: no standard price lists a cached-input discount. Ministral 3 8B from USD 0.015 per 1M cached input tokens (Mistral AI; 1 of 2 standard prices list a discount). Cached tokens are billed at the input price where no cache price is published.
Which has the larger context window, Llama 4 Scout or Ministral 3 8B?
Llama 4 Scout: up to 1,048,576 tokens, against 262,144 for Ministral 3 8B, as stated on providers' pricing pages. Some providers offer less than the maximum; check the model pages before you pick the cheapest row for long prompts.