feat(deepinfra): update model YAMLs [bot]

[models-bot]

Auto-generated by poc-agent for provider deepinfra.

---

Note

Low Risk
Configuration-only provider catalog changes with no runtime code; incorrect pricing or feature flags could mislead routing/cost estimates but scope is limited to metadata files.

Overview
Bot-synced DeepInfra provider model YAMLs to match current catalog data.

google/gemma-4-31B-it-turbo pricing is revised (lower input/output token costs) and cache_read_input_token_cost is added; json_output is dropped from features.

Meta-Llama-3.1-8B-Instruct and Nemotron-3-Nano-Omni-30B-A3B-Reasoning no longer include deprecationDate, aligning them with active listings rather than a July 2026 deprecation.

ByteDance/Seed-2.0-code only removes stale comments about structured/JSON output not being supported; capabilities in the file are unchanged.

^{Reviewed by Cursor Bugbot for commit 45e7f9f. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

/test-models

[harshiv-26]

Gateway test results

• Total: 12
• Passed: 8
• Failed: 1
• Validation failed: 1
• Errored: 0
• Skipped: 2
• Success rate: 80.0%

Provider	Model	Scenarios
deepinfra	ByteDance/Seed-2.0-code	success: tool-call:stream, tool-call, params, params:stream failure: json-output validation_failure: json-output:stream
deepinfra	google/gemma-4-31B-it-turbo	success: params:stream, tool-call:stream, params, tool-call
deepinfra	meta-llama/Meta-Llama-3.1-8B-Instruct	skipped: skip-check
deepinfra	nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning	skipped: skip-check

Failures (2)

deepinfra/ByteDance/Seed-2.0-code — json-output:stream (validation_failure)

Error

Traceback (most recent call last):
  File "/tmp/tmp6eobnhfv/snippet.py", line 24, in <module>
    raise Exception("VALIDATION FAILED: json-output stream - no content received")
Exception: VALIDATION FAILED: json-output stream - no content received

Code snippet

from 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/ByteDance-Seed-2.0-code",
    messages=[
        {"role": "user", "content": "List 3 colors with their hex codes in JSON."},
    ],
    response_format={"type": "json_object"},
    stream=True,
)
import json as _json

_accumulated = ""
for chunk in response:
    if chunk.choices and len(chunk.choices) > 0:
        delta = chunk.choices[0].delta
        if delta.content is not None:
            _accumulated += delta.content
            print(delta.content, end="", flush=True)

if not _accumulated:
    raise Exception("VALIDATION FAILED: json-output stream - no content received")

_json.loads(_accumulated)
print("\nVALIDATION: json-output stream SUCCESS")

deepinfra/ByteDance/Seed-2.0-code — json-output (failure)

Error

Traceback (most recent call last):
  File "/tmp/tmpvg7ejga0/snippet.py", line 5, in <module>
    response = client.chat.completions.create(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/openai/_utils/_utils.py", line 286, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/openai/resources/chat/completions/completions.py", line 1147, in create
    return self._post(
           ^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/openai/_base_client.py", line 1259, in post
    return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/openai/_base_client.py", line 1047, in request
    raise self._make_status_error_from_response(err.response) from None
openai.InternalServerError: Error code: 500 - {'status': 'failure', 'message': 'Invalid response received from deepinfra: {"error":{"message":"{\\"error\\":{\\"code\\":\\"InvalidParameter\\",\\"message\\":\\"The parameter `response_format.type` specified in the request are not valid: `json_object` is not supported by this model. Request id: 02178615841534011300f35b2a874e35aceb4d35afb57b86f9d82\\",\\"param\\":\\"response_format.type\\",\\"type\\":\\"BadRequest\\"}}","type":"api_error","param":null,"code":null}}', 'error': {'message': 'Invalid response received from deepinfra: {"error":{"message":"{\\"error\\":{\\"code\\":\\"InvalidParameter\\",\\"message\\":\\"The parameter `response_format.type` specified in the request are not valid: `json_object` is not supported by this model. Request id: 02178615841534011300f35b2a874e35aceb4d35afb57b86f9d82\\",\\"param\\":\\"response_format.type\\",\\"type\\":\\"BadRequest\\"}}","type":"api_error","param":null,"code":null}}', 'type': 'APIError', 'code': '500'}, 'error_origin_level': 'api_error', 'provider': 'deepinfra'}

Code snippet

from 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/ByteDance-Seed-2.0-code",
    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")

Successes (8)

deepinfra/google/gemma-4-31B-it-turbo — params:stream (success)

Output

The capital of France is Paris.

deepinfra/google/gemma-4-31B-it-turbo — tool-call:stream (success)

Output

{"location": "London"}
VALIDATION: tool-call stream SUCCESS

deepinfra/google/gemma-4-31B-it-turbo — params (success)

Output

The capital of France is Paris.

deepinfra/google/gemma-4-31B-it-turbo — tool-call (success)

Output

Function: get_weather
Arguments: {"location": "London"}
VALIDATION: tool-call SUCCESS

deepinfra/ByteDance/Seed-2.0-code — tool-call:stream (success)

Output

{"location": "London"}
VALIDATION: tool-call stream SUCCESS

deepinfra/ByteDance/Seed-2.0-code — tool-call (success)

Output

Function: get_weather
Arguments: {"location": "London"}
VALIDATION: tool-call SUCCESS

deepinfra/ByteDance/Seed-2.0-code — params (success)

Output

The capital of France is **Paris**.

Paris is also France’s most populous city, a global cultural and economic hub, and sits along the Seine River in 
... (truncated, 263 chars omitted)

deepinfra/ByteDance/Seed-2.0-code — params:stream (success)

Output

The capital of France is Paris. Paris is also France's most populous city, political, cultural, and economic hub, located along the Seine River in the
... (truncated, 18234 chars omitted)

Skipped (2)

deepinfra/meta-llama/Meta-Llama-3.1-8B-Instruct — skip-check (skipped)

Skip reason

deprecated or retired model

deepinfra/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning — skip-check (skipped)

Skip reason

deprecated or retired model

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 2 potential issues.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 562a237. Configure here.}

[cursor]

Incorrect gemma turbo pricing

Medium Severity

Input/output costs were changed to 9e-8 / 3.4e-7 with a new cache-read rate, but DeepInfra still publishes standard turbo pricing as $0.12 / $0.37 per 1M tokens with no listed cache rate. Downstream billing based on this YAML would be wrong.

 

^{Reviewed by Cursor Bugbot for commit 562a237. Configure here.}

[Archit-Kumar-16]

remove this

[LordGameleo]

Seed-2.0-code: json_output was added to features, but a comment two lines below still says "json_output is not supported". Contradictory — please clarify which is correct.

[github-actions]

/test-models

[github-actions]

/test-models

[github-actions]

/test-models