Fireworks AI model pricing.

Prices checked

Fireworks AI sells 52 prices on 53 models, from $0.09 per million tokens, checked .

Every model Fireworks AI sells

  1. 1

    Nemotron 3.5 Lightning

    Runs it themselves

    $0.05 in · $0.20 out

    $0.09/M tokens
    View
  2. 2

    Ministral 3 3B

    Same as maker

    $0.10 in · $0.10 out

    $0.10/M tokens
    View
  3. 3

    gpt-oss-20b

    Costs more

    $0.07 in · $0.30 out

    $0.13/M tokens
    View
  4. 4

    $0.14 in · $0.28 out

    $0.18/M tokens
    View
  5. 5

    Llama Guard 3 8B

    Runs it themselves

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  6. 6

    Ministral 3 14B

    Same as maker

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  7. 7

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  8. 8

    Mistral Nemo

    Costs more

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  9. 9

    MythoMax 13B

    Costs more

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  10. 10

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  11. 11

    Qwen3 14B

    Runs it themselves

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  12. 12

    Qwen3 8B

    Costs more

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  13. 13

    Qwen3 VL 8B Instruct

    Runs it themselves

    $0.20 in · $0.20 out

    $0.20/M tokens
    View
  14. 14

    GLM-5.3-Flash

    Same as maker

    $0.15 in · $0.50 out

    $0.24/M tokens
    View
  15. 15

    gpt-oss-120b

    Same as maker

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  16. 16

    MiniMax-M3.1-Flash

    Same as maker

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  17. 17

    Qwen3 30B A3B

    Costs more

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  18. 18

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  19. 19

    Qwen3-VL 30B-A3B

    Runs it themselves

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  20. 20

    Qwen3 VL 30B A3B Thinking

    Runs it themselves

    $0.15 in · $0.60 out

    $0.26/M tokens
    View
  21. 21

    $0.22 in · $0.66 out

    $0.33/M tokens
    View
  22. 22

    Qwen3 235B-A22B

    Runs it themselves

    $0.22 in · $0.88 out

    $0.39/M tokens
    View
  23. 23

    $0.22 in · $0.88 out

    $0.39/M tokens
    View
  24. 24

    Qwen3-VL 235B-A22B

    Runs it themselves

    $0.22 in · $0.88 out

    $0.39/M tokens
    View
  25. 25

    Qwen3 VL 235B A22B Thinking

    Runs it themselves

    $0.22 in · $0.88 out

    $0.39/M tokens
    View

Where to start

Kind

Host

based in US

Free tier

No

every price is paid

What Fireworks AI charges per million tokens embedded

Priced on input alone, since an embedding model writes no output, so never held beside a chat model's blended price. Every embedding model

Embedding models, per 1M input tokens

  1. 1

    Qwen3 Embedding 8B

    No maker price

    Alibaba Cloud

    $0.10/M tokens

Frequently asked

What is Fireworks AI?
Serves open-weight models on its own GPUs; on-demand deployments are priced per GPU-hour. It sells 53 of the models we track, and trades as Fireworks AI, Inc., based in US.
Is Fireworks AI cheaper than buying from the maker?
19 of its 52 prices are below the maker's own price, and 15 of those come with a published reason, so they can lead a ranking.
Does Fireworks AI train on what you send it?
Fireworks AI does not say whether it trains on what you send it. It does not say how long it keeps requests.
How current are these Fireworks AI prices?
They come from the public catalogs we read, read every 3 days; the reading on this page is from 31 Oct 2026. Prices exclude tax.

Models it sells

Other hosts

Where to go next