Fix: Add claude-opus-4-7 to NO_SUPPORT_TEMPERATURE_MODELS (#2400)#2409
Open
utsab345 wants to merge 1 commit into
Open
Fix: Add claude-opus-4-7 to NO_SUPPORT_TEMPERATURE_MODELS (#2400)#2409utsab345 wants to merge 1 commit into
utsab345 wants to merge 1 commit into
Conversation
Contributor
Review Summary by QodoAdd claude-opus-4-7 variants to NO_SUPPORT_TEMPERATURE_MODELS
WalkthroughsDescription• Add Claude Opus 4-7 model variants to temperature exclusion list • Fixes 400 error from Anthropic deprecating temperature parameter • Covers all provider variants: Vertex AI, Anthropic, Bedrock • Follows existing pattern for unsupported temperature models Diagramflowchart LR
A["Claude Opus 4-7 Models"] -->|"Add to exclusion list"| B["NO_SUPPORT_TEMPERATURE_MODELS"]
B -->|"LiteLLMAIHandler skips"| C["Temperature Parameter"]
C -->|"Prevents"| D["400 BadRequestError"]
File Changes1. pr_agent/algo/__init__.py
|
Contributor
Code Review by Qodo
Context used 1. Model lists inconsistent
|
![qodo-free-for-open-source-projects[bot]](https://avatars.githubusercontent.com/in/1068000?s=60&v=4){kind=small}
Comment on lines
+299
to
+304
| "claude-3-7-opus", | ||
| "claude-opus-4.7", | ||
| "vertex_ai/claude-opus-4-7", | ||
| "anthropic/claude-opus-4-7", | ||
| "bedrock/anthropic.claude-opus-4-7", | ||
| "bedrock/anthropic.claude-opus-4-7-v1:0", |
Contributor
There was a problem hiding this comment.
1. Model lists inconsistent 🐞 Bug ≡ Correctness
NO_SUPPORT_TEMPERATURE_MODELS now includes model IDs that are not present in MAX_TOKENS, so configuring those models will still crash in get_max_tokens() (even though temperature is skipped). This specifically affects bedrock/anthropic.claude-opus-4-7-v1:0 and the newly-added aliases claude-3-7-opus and claude-opus-4.7.
Agent Prompt
## Issue description
Several newly added entries in `NO_SUPPORT_TEMPERATURE_MODELS` are not recognized by the rest of the system because they are missing from `MAX_TOKENS`. When a user configures one of these models, PR-Agent will raise an exception in `get_max_tokens()` before making any LLM request.
## Issue Context
- `get_max_tokens(model)` hard-fails when `model` is not in `MAX_TOKENS` and no `config.custom_model_max_tokens` is set.
- The PR added `bedrock/anthropic.claude-opus-4-7-v1:0` (and some additional aliases) to `NO_SUPPORT_TEMPERATURE_MODELS`, but did not add corresponding `MAX_TOKENS` entries.
## Fix Focus Areas
- pr_agent/algo/__init__.py[278-307]
- pr_agent/algo/__init__.py[150-160]
- pr_agent/algo/utils.py[992-1010]
## Suggested fix
1. Add missing `MAX_TOKENS` entries for the newly-added model IDs that are intended to be supported (at minimum `bedrock/anthropic.claude-opus-4-7-v1:0` should map to the same token limit as `bedrock/anthropic.claude-opus-4-7`).
2. For `claude-3-7-opus` and `claude-opus-4.7`, either:
- add them to `MAX_TOKENS` with the correct limits if they are valid model names in this project, or
- remove them from `NO_SUPPORT_TEMPERATURE_MODELS` if they are typos / not supported.
3. (Optional but recommended) Add/extend a unit test similar to `test_claude_opus_4_7_model_max_tokens` to cover the `-v1:0` Bedrock variant if you keep it.
ⓘ Copy this prompt and use it to remediate the issue with your preferred AI generation tools
Author
There was a problem hiding this comment.
Thank you for the review. The issue about inconsistent model lists has been resolved.
The missing models have been added to MAX_TOKENS in commit 80aa7ac:
claude-3-7-opus: 200,000 tokensclaude-opus-4.7: 1,000,000 tokensbedrock/anthropic.claude-opus-4-7-v1:0: 1,000,000 tokens
All Claude Opus 4-7 variants are now present in both MAX_TOKENS (for token calculation) and NO_SUPPORT_TEMPERATURE_MODELS (to skip temperature parameter). This ensures get_max_tokens() works correctly when these models are configured.
The token limits are based on Anthropic's official documentation:
- Claude Opus 4-7: 1,000,000 token context window
- Claude 3.7 Opus: 200,000 token context window
The change maintains consistency between both lists and follows the existing pattern for model configuration.
Please review the updated PR. Thank you.
utsab345
force-pushed
the
fix/claude-opus-4-7-temperature
branch
from
e5460ca to
80aa7ac
Compare
Contributor
|
Persistent review updated to latest commit 80aa7ac |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fix: Add claude-opus-4-7 to NO_SUPPORT_TEMPERATURE_MODELS
Problem Description
Fixes issue #2400
When using Claude Opus 4-7 models (via Vertex AI, Anthropic, or Bedrock), PR-Agent fails with a 400 error because Anthropic has deprecated the
temperatureparameter for these models.Error message: itellm.BadRequestError: Vertex_aiException BadRequestError -
{"type":"error","error":{"type":"invalid_request_error","message":"temperature is deprecated for this model."}}
Affected models:
Solution
Added all Claude Opus 4-7 model variants to the existing
NO_SUPPORT_TEMPERATURE_MODELSlist inpr_agent/algo/__init__.py.This follows the same pattern used for other models that don't support temperature (o1, o3, gpt-5-mini, etc.). The
LiteLLMAIHandleralready has logic to skip sending the temperature parameter for any model in this list.Changes Made
File changed:
pr_agent/algo/__init__.pyAdded to
NO_SUPPORT_TEMPERATURE_MODELS: