Llama 3.1 8B Instruct vs Mistral Small 3.

model vs model · prices as of

Llama 3.1 8B Instruct takes 2 of 5 rounds against Mistral Small 3. It is also the cheaper of the two to run, at $0.10 per million tokens against $0.23.

Metawins 2 rounds

Llama 3.1 8B Instruct

$0.10/M tokens

cheapest we can explain, at Kilo Gateway · $0.0144 per 1,000 pages

131k context · tool calling · open weights

Mistral AI

Mistral Small 3

$0.23/M tokens

OpenRouter list price, nothing below it is accounted for · $0.0348 per 1,000 pages

33k context · open weights

List price in (lower wins)

$0.05/M

$0.05/M

List price out (lower wins)

$0.08/M

$0.08/M

Cheapest we can explain (lower wins)

$0.10/M at Kilo Gateway

none below list

Standard price (lower wins)

$0.23/M

$0.23/M

Cost per 1,000 pages (lower wins)

$0.0144

$0.0348

Context

131k tokens

33k tokens

Open weights

yes

yes

Tool calling

yes

no

Structured output

yes

yes

Reads images

no

no

Reasoning

no

no

Released

23 Jul 2024

30 Jan 2025

Providers selling it

24

10

Price we measure against

OpenRouter

OpenRouter

Trains on your prompts (no is better)

not stated

no provider to check

Providers selling both · 5

  • DeepInfra$0.10 / $0.23
  • Kilo Gateway$0.10 / $0.23
  • TrustedRouter$0.12 / $0.24
  • OpenRouter$0.23 / $0.23
  • NanoGPT$0.25 / $0.69

Llama 3.1 8B Instruct then Mistral Small 3, per million tokens

Go further

Add a third or fourth model, or switch the weighting, in the tool.

What moved in the last seven days

  1. Llama 3.1 8B InstructNous Portal$0.02 / $0.03 $0.02 / $0.0425% dearer ·

Frequently asked

Which is cheaper, Llama 3.1 8B Instruct or Mistral Small 3?
Llama 3.1 8B Instruct, at $0.10 per million tokens against $0.23 for Mistral Small 3, three parts input to one part output, read on 16 Sept 2026.
Where is each one cheapest?
Llama 3.1 8B Instruct: $0.10 at Kilo Gateway. Mistral Small 3: nobody goes below OpenRouter's own price with a reason we can point at, so $0.23 is the number to beat.
Can I buy both from one provider?
Yes. 5 providers sell both: DeepInfra, Kilo Gateway, TrustedRouter, OpenRouter, NanoGPT. The cheapest of them for the two together is DeepInfra.
Which holds more context?
Llama 3.1 8B Instruct, at 131k tokens against 33k.

Prices exclude tax and are per million tokens, read 16 Sept 2026 from each provider's own catalog. The standard price blends three parts input to one part output; a page is 600 tokens in and 60 out. Bars are scaled to the larger value in each row.

Related head-to-headsgpt-oss-20b vs Llama 3.1 8B Instructgpt-oss-20b vs Mistral Small 3Llama 3.1 8B Instruct vs Nova Micro 1.0Llama 3.1 8B Instruct vs Qwen3.7 FlashMorecompare any two modelsevery model we track