Llama 4 Scout pricing.

Prices checked

Llama 4 Scout costs $0.10 in and $0.30 out per million tokens from OpenRouter; the cheapest price is $0.18 at NanoGPT, checked .

OpenRouter's own price

OpenRouter

$0.15/M tokens

$0.10 in · $0.30 out

83% below the typical price of the 399 text models we track.

Cheapest price

NanoGPT

$0.18/M tokens

$0.09 in · $0.46 out

Nobody sells it for less than OpenRouter with a reason we can point at.

tool callingreads imagesopen weights128k contextreleased 5 Apr 2025

Every provider selling Llama 4 Scout

  1. 1

    Lambda

    Runs it themselves

    $0.05 in · $0.10 out · 16k context here

    $0.06/M tokens
  2. 2

    $0.08 in · $0.24 out · 328k context here

    $0.12/M tokens
  3. 3

    Helicone

    Runs it themselves

    $0.08 in · $0.30 out · 131k context here

    $0.14/M tokens
  4. 4

    Nscale

    Resells

    $0.09 in · $0.29 out · 890k context here

    $0.14/M tokens
  5. 5

    OpenRouter

    Same as maker

    $0.10 in · $0.30 out · 1.3M context here

    $0.15/M tokens
  6. 6

    DeepInfra

    Same as maker

    $0.10 in · $0.30 out · 328k context here

    $0.15/M tokens
  7. 7

    DeepInfra

    Same as maker

    $0.10 in · $0.30 out · 328k context here

    $0.15/M tokens
  8. 8

    Nous Portal

    Same as maker

    $0.10 in · $0.30 out · 1.3M context here

    $0.15/M tokens
  9. 9

    TrustedRouter

    Costs more

    $0.11 in · $0.32 out · 131k context here

    $0.16/M tokens
  10. 10

    NanoGPT

    Resells

    $0.09 in · $0.46 out · 328k context here

    $0.18/M tokens
  11. 11

    Jiekou

    Same as maker

    $0.10 in · $0.50 out · 131k context here

    $0.20/M tokens
  12. 12

    Novita

    Costs more

    $0.18 in · $0.59 out · 131k context here

    $0.28/M tokens
  13. 13

    Together AI

    Costs more

    $0.18 in · $0.59 out · 1M context here

    $0.28/M tokens
  14. 14

    $0.17 in · $0.66 out

    $0.29/M tokens
  15. 15

    FastRouter

    Costs more

    $0.17 in · $0.66 out · 33k context here

    $0.29/M tokens
  16. 16

    $0.20 in · $0.78 out

    $0.35/M tokens
  17. 17

    $0.20 in · $0.78 out

    $0.35/M tokens
  18. 18

    Google

    Costs more

    $0.25 in · $0.70 out · 1.3M context here

    $0.36/M tokens
  19. 19

    $0.27 in · $0.85 out · 131k context here

    $0.42/M tokens
  20. 20

    Eden AI

    Costs more

    $0.27 in · $0.85 out · 131k context here

    $0.42/M tokens

Not ranked

Below the maker's own price with nothing published to account for it, from a catalog we could not read reliably, or a free tier, which is a limit rather than a price. Why

  1. Meta

    Free tier

    given away with limits

    free
What it costs for your work
pages a month
$0.08

A month at OpenRouter's own price

$0.08

A month at NanoGPT, the cheapest

under $0.01

What you pay more

A page is about 600 tokens in and 60 out, a support reply 200 in and 300 out, an email 150 in and 250 out. Nothing else is counted: no cache discount, no batch rate, no free tier.

Run it yourself

On a server by the month

Meta publishes the weights, so you can run Llama 4 Scout on a server you rent by the month. It needs about 68.4 GB of RAM on a CPU, and the cheapest plan we track with that much and a disk included is Linode Linode 90GB at $240 a month, streaming about 0.7–1.4 tok/s. Every plan that fits

On a GPU by the hour

The GH200 holds Llama 4 Scout in 96 GB of memory against the 68.4 GB the weights, the context and the runtime need. The cheapest one we track is Vultr's vbm-72c-480gb-gh200-gpu at $1.99/GPU-hr, which is $1,452.70 for the 730 hours in a month, checked 21 Oct. Every price for the GH200

How we count

Prices are per million tokens. The big number on each row is blended 3 in : 1 out, the way most work uses a model. A price below OpenRouter's own is listed as cheapest only when the provider says why. Meta trained Llama 4 Scout and sells it on no catalog we read, so the price every row is measured against is OpenRouter's.

Runs it themselves
The provider runs the open weights on hardware of its own, so the price is its own rather than a resale.
Resells
The provider sells another company's endpoint, at a margin of its own or at the maker's price.
Same as maker
Sold by someone else at the price the maker charges.
Costs more
Sold by someone else for more than the maker charges.
Free tier
Given away inside a quota. A limit is not a price, so it is never called cheapest.

Similar models

Models from other labs that do what Llama 4 Scout does, cheapest first. Each percentage is against Llama 4 Scout's cheapest price, $0.15 per M tokens.

Frequently asked

Who sells Llama 4 Scout cheapest?
NanoGPT, at $0.09 in and $0.46 out per million tokens, checked 21 Oct 2026. The provider sells another company's endpoint, at a margin of its own or at the maker's price.
Is Llama 4 Scout open weights?
Yes. Meta publishes the weights, so anyone may host Llama 4 Scout, which is why 19 providers sell it and why their prices differ so much.
How much context does Llama 4 Scout hold?
128,000 tokens in one call, which is about 128k, and it can call your own functions.
How does Llama 4 Scout compare with the rest of the market?
At $0.15 per million tokens, OpenRouter's own price is 83% below the typical $0.89 across the 399 text models we track.

More from Meta