fix: update sonnet 4 configs to reflect million-token context window pricing #14639
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
| "cache_read_input_token_cost": 3e-07, | ||
| "input_cost_per_token": 3e-06, | ||
| "input_cost_per_token_above_200k_tokens": 6e-06, | ||
| "output_cost_per_token_above_200k_tokens": 2.25e-05, |
There was a problem hiding this comment.
where are you seeing this? @danielmklein
There was a problem hiding this comment.
@krrishdholakia This pricing specifically? It's the "long context pricing" described here. Let me know if that's not what you're asking about!
There was a problem hiding this comment.
hey @danielmklein this pricing page is for the anthropic api. But the change you're suggesting is for bedrock.
Can you confirm the bedrock pricing is the same as that of the anthropic api in this case?
There was a problem hiding this comment.
@krrishdholakia Yes, the bedrock pricing page for long context Sonnet 4 matches these figures: https://aws.amazon.com/bedrock/pricing/ (unfortunately they don't support linking directly to the relevant table)
Title
Update Sonnet 4 configurations to support million-token context windows (with correct pricing)
Relevant issues
#13724
#13759
Pre-Submission checklist
Please complete all items before asking a LiteLLM maintainer to review your PR
tests/litellm/directory, Adding at least 1 test is a hard requirement - see detailsmake test-unitType
🐛 Bug Fix
Changes
This updates the relevant elements in
model_prices_and_context_window.jsonandlitellm/model_prices_and_context_window_backup.jsonto reflect the token limits and pricing tiers for long-context Sonnet 4.