feat(deepinfra/anthropic/claude-sonnet-5): add new models [bot]#1706
feat(deepinfra/anthropic/claude-sonnet-5): add new models [bot]#1706models-bot[bot] wants to merge 2 commits into
Conversation
![cursor[bot]](https://avatars.githubusercontent.com/in/1210556?s=60&v=4){kind=small}
|
/test-models |
There was a problem hiding this comment.
Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.
❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.
Reviewed by Cursor Bugbot for commit 1a0c7e5. Configure here.
| - json_output | ||
| limits: | ||
| context_window: 1000000 | ||
| max_tokens: 1000000 |
There was a problem hiding this comment.
Wrong max output token limit
Medium Severity
limits.max_tokens is set to 1000000, matching context_window instead of Claude Sonnet 5’s actual maximum response size. In this registry, max_tokens is the cap on generated tokens; Sonnet 5 supports a 1M context but only about 128K output, as in the canonical Anthropic and OpenRouter entries.
Reviewed by Cursor Bugbot for commit 1a0c7e5. Configure here.
Gateway test results
Failures (4)
ErrorCode snippetfrom openai import OpenAI
client = OpenAI(api_key="***", base_url="https://internal.devtest.truefoundry.tech/api/llm")
response = client.chat.completions.create(
model="test-v2-deepinfra/anthropic-claude-sonnet-5",
messages=[
{"role": "user", "content": "How to calculate 3^3^3^3? Think step by step and show all reasoning."},
],
reasoning_effort="medium",
stream=True,
)
_reasoning_detected = False
for chunk in response:
if chunk.choices and len(chunk.choices) > 0:
delta = chunk.choices[0].delta
if delta.content is not None:
print(delta.content, end="", flush=True)
if getattr(delta, "reasoning_content", None) is not None:
_reasoning_detected = True
if getattr(delta, "reasoning", None) is not None:
_reasoning_detected = True
_usage = getattr(chunk, "usage", None)
if _usage is not None:
_details = getattr(_usage, "completion_tokens_details", None)
if _details and getattr(_details, "reasoning_tokens", 0) > 0:
_reasoning_detected = True
if not _reasoning_detected:
raise Exception("VALIDATION FAILED: reasoning stream - no reasoning information in stream")
print("\nVALIDATION: reasoning stream SUCCESS")
ErrorCode snippetfrom openai import OpenAI
client = OpenAI(api_key="***", base_url="https://internal.devtest.truefoundry.tech/api/llm")
response = client.chat.completions.create(
model="test-v2-deepinfra/anthropic-claude-sonnet-5",
messages=[
{"role": "user", "content": "List 3 colors with their hex codes in JSON."},
],
response_format={"type": "json_object"},
stream=False,
)
import json as _json
_content = response.choices[0].message.content
print(_content)
if not _content:
raise Exception("VALIDATION FAILED: json-output - response content is empty")
_json.loads(_content)
print("VALIDATION: json-output SUCCESS")OutputTraceback (most recent call last): OutputTraceback (most recent call last): Successes (4)
Output
Output
Output
Output |
1 participant
Auto-generated by model-addition-agent for
deepinfra/anthropic/claude-sonnet-5.Note
Low Risk
Adds catalog metadata only; no runtime, auth, or routing logic changes.
Overview
Adds a new DeepInfra provider YAML for
anthropic/claude-sonnet-5, registering the model as active serverless chat with the same capabilities as the existing Sonnet entries (function calling, JSON output, text/image in, 1M context, thinking enabled).Pricing is set to $2e-6 / $1e-5 per input/output token (globally via
region: "*"), with a source link to DeepInfra’s model page.Reviewed by Cursor Bugbot for commit 1a0c7e5. Bugbot is set up for automated code reviews on this repo. Configure here.