Feat/new models #248
Feat/new models #248
Conversation
# Conflicts: # packages/gnosis/customs/ofv_market_resolver/component.yaml # packages/packages.json
| @@ -348,6 +348,11 @@ def error_response(msg: str) -> Tuple[str, None, None, None]: | |||
| "limit_max_tokens": 8192, | |||
| "temperature": 0, | |||
| }, | |||
| "gpt-4o-2024-05-13": { | |||
| "default_max_tokens": 500, | |||
There was a problem hiding this comment.
I just copied over from above.
| @@ -348,6 +348,11 @@ def error_response(msg: str) -> Tuple[str, None, None, None]: | |||
| "limit_max_tokens": 8192, | |||
| "temperature": 0, | |||
| }, | |||
| "gpt-4o-2024-05-13": { | |||
There was a problem hiding this comment.
The "gpt-4o-2024-08-06" model is cheaper...?
There was a problem hiding this comment.
Thanks, will replace.
| @@ -254,6 +254,11 @@ def embeddings(self, model, input): | |||
| "limit_max_tokens": 8192, | |||
| "temperature": 0, | |||
| }, | |||
| "gpt-4o-2024-05-13": { | |||
| "default_max_tokens": 500, | |||
| "limit_max_tokens": 4096, | |||
There was a problem hiding this comment.
limit max tokens = 128_000
you can check it here
https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json
There was a problem hiding this comment.
I think thats for the input no?
There was a problem hiding this comment.
in the url that I shared.... search for sonnet and check the config that is there and the one you wrote for limit_max_tokens you put the max input tokens value, not the max_output tokens value. Then why for the gpt model you are putting in limit_max_tokens the max output tokens instead of the input max?
| }, | ||
| "gpt-4o-2024-05-13": { | ||
| "default_max_tokens": 500, | ||
| "limit_max_tokens": 4096, |
There was a problem hiding this comment.
limit max tokens = 128_000
you can check it here
https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json
| @@ -348,6 +348,11 @@ def error_response(msg: str) -> Tuple[str, None, None, None]: | |||
| "limit_max_tokens": 8192, | |||
| "temperature": 0, | |||
| }, | |||
| "gpt-4o-2024-05-13": { | |||
| "default_max_tokens": 500, | |||
| "limit_max_tokens": 4096, | |||
There was a problem hiding this comment.
limit max tokens = 128_000
you can check it here
https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json
| @@ -36,9 +36,11 @@ class TokenCounterCallback: | |||
| "gpt-4-turbo-preview": {"input": 0.01, "output": 0.03}, | |||
| "gpt-4-0125-preview": {"input": 0.01, "output": 0.03}, | |||
| "gpt-4-1106-preview": {"input": 0.01, "output": 0.03}, | |||
| "gpt-4o-2024-08-06": {"input": 0.01, "output": 0.03}, | |||
There was a problem hiding this comment.
where did you get those values? they are different from the ones here
https://raw.githubusercontent.com/BerriAI/litellm/main/model_prices_and_context_window.json
Proposed changes
This PR adds support for
gpt4oandclaude3.5-sonnet.Types of changes
What types of changes does your code introduce? (A breaking change is a fix or feature that would cause existing functionality and APIs to not work as expected.)
Put an
xin the box that appliesChecklist
Put an
xin the boxes that apply.mainbranch (left side). Also you should start your branch off ourmain.