4 ms·

No pricing, but given that OpenAI's GPT-4 doubles the cost-per-token if you go from 8k to a 32k context window, I suspect the pricing here will be 2-4x from the

No pricing, but given that OpenAI's GPT-4 doubles the cost-per-token if you go from 8k to a 32k context window, I suspect the pricing here will be 2-4x from the base Claude model which is 9k: https://cdn2.assets-servd.host/anthropic-website/production/images/apr-pricing-tokens_2023-05-10-213035_fhnc.pdf https://cdn2.assets-servd.host/anthropic-website/production/... Although with flash attention, who knows if marginal cost scales that consistently.

adamkochanowicz 4y ago
https://cdn2.assets-servd.host/anthropic-website/production/images/apr-pricing-tokens.pdf https://cdn2.assets-servd.host/anthropic-website/production/...
- minimaxir 4y ago
  Those are the same SKUs I linked. The new model are a different model identified that's not listed in the pricing doc, although it sounds like the intent may be to replace the base from looking at the API docs: https://console.anthropic.com/docs/api/reference#-v1-complete https://console.anthropic.com/docs/api/reference#-v1-complet...
f_devd 4y ago
<4x would be quite optimistic, at ~11x the tokens the amount of compute/memory required would be n^10 (even with the lower starting point of flash attention) so unless they are already have excessive margins it wouldn't make much sense to go that low.
- sp332 4y ago
  I was assuming they used a different architecture to get the increase instead of just letting it eat hardware that way. Especially with the speed numbers in the post.
l1n 4y ago
Pricing is the same as the base model.
- jimsimmons 4y ago
  Confirmation here: https://twitter.com/AnthropicAI/status/1656743460769259521?s=20 https://twitter.com/AnthropicAI/status/1656743460769259521?s...
  - minimaxir 4y ago
    Huh. Well that changes things.
    - rat9988 4y ago
      Only for the duration of the beta
      - jimsimmons 4y ago
        Source?
        felixgallo 4y ago
        the actual tweet you linked.
        jimsimmons 4y ago
        It doesn’t say exclusively for the beta period
        scoopertrooper 4y ago
        With an extremely literal reading you are correct, but there was clearly an implication.

adamkochanowicz 4y ago

https://cdn2.assets-servd.host/anthropic-website/production/images/apr-pricing-tokens.pdf https://cdn2.assets-servd.host/anthropic-website/production/...

minimaxir 4y ago
Those are the same SKUs I linked. The new model are a different model identified that's not listed in the pricing doc, although it sounds like the intent may be to replace the base from looking at the API docs: https://console.anthropic.com/docs/api/reference#-v1-complete https://console.anthropic.com/docs/api/reference#-v1-complet...

f_devd 4y ago

<4x would be quite optimistic, at ~11x the tokens the amount of compute/memory required would be n^10 (even with the lower starting point of flash attention) so unless they are already have excessive margins it wouldn't make much sense to go that low.

sp332 4y ago
I was assuming they used a different architecture to get the increase instead of just letting it eat hardware that way. Especially with the speed numbers in the post.

l1n 4y ago

Pricing is the same as the base model.

jimsimmons 4y ago
Confirmation here: https://twitter.com/AnthropicAI/status/1656743460769259521?s=20 https://twitter.com/AnthropicAI/status/1656743460769259521?s...
- minimaxir 4y ago
  Huh. Well that changes things.
  - rat9988 4y ago
    Only for the duration of the beta
    - jimsimmons 4y ago
      Source?
      - felixgallo 4y ago
        the actual tweet you linked.
        jimsimmons 4y ago
        It doesn’t say exclusively for the beta period
        scoopertrooper 4y ago
        With an extremely literal reading you are correct, but there was clearly an implication.