[DO NOT MERGE]: test cohere new

[harshiv-26]

Note

Low Risk
Low risk: a single config value type/format change with no logic modifications; only potential impact is on consumers that strictly parse numeric types.

Overview
Normalizes the Cohere model config in providers/cohere/command-a-plus-05-2026.yaml by changing costs.input_cost_per_token from 0 to 0.0.

^{Reviewed by Cursor Bugbot for commit 4faa883. Bugbot is set up for automated code reviews on this repo. Configure here.}

[harshiv-26]

/test-models

[harshiv-26]

Gateway test results

• Total: 8
• Passed: 6
• Failed: 0
• Validation failed: 2
• Errored: 0
• Skipped: 0
• Success rate: 75.0%

Provider	Model	Scenarios
cohere	command-a-plus-05-2026	success: params:stream, params, tool-call, tool-call:stream, structured-output, structured-output:stream validation_failure: reasoning:stream, reasoning

Failures (2)

cohere/command-a-plus-05-2026 — reasoning:stream (validation_failure)

Error:

Traceback (most recent call last):
  File "/tmp/tmpqacaou3e/snippet.py", line 35, in <module>
    raise Exception("VALIDATION FAILED: reasoning stream - no reasoning information in stream")
Exception: VALIDATION FAILED: reasoning stream - no reasoning information in stream

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-cohere/command-a-plus-05-2026",
    messages=[
        {"role": "system", "content": "You are a helpful assistant. You MUST think step by step and show your reasoning. Never skip reasoning steps."},
        {"role": "user", "content": "Hi"},
        {"role": "assistant", "content": "Hi, how can I help you"},
        {"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")

cohere/command-a-plus-05-2026 — reasoning (validation_failure)

Error:

Traceback (most recent call last):
  File "/tmp/tmpggdw_0r0/snippet.py", line 43, in <module>
    raise Exception("VALIDATION FAILED: reasoning - no reasoning information in response")
Exception: VALIDATION FAILED: reasoning - no reasoning information in response

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-cohere/command-a-plus-05-2026",
    messages=[
        {"role": "system", "content": "You are a helpful assistant. You MUST think step by step and show your reasoning. Never skip reasoning steps."},
        {"role": "user", "content": "Hi"},
        {"role": "assistant", "content": "Hi, how can I help you"},
        {"role": "user", "content": "How to calculate 3^3^3^3? Think step by step and show all reasoning."},
    ],
    reasoning_effort="medium",
    stream=False,
)

_usage = getattr(response, "usage", None)
_reasoning_detected = False

_choices = getattr(response, "choices", None)
if _choices and len(_choices) > 0:
    _message = getattr(_choices[0], "message", None)
else:
    _message = None

if _message and getattr(_message, "content", None) is not None:
    print(_message.content)

if _usage is not None:
    _output_token_details = getattr(_usage, "completion_tokens_details", None)
    if _output_token_details and getattr(_output_token_details, "reasoning_tokens", 0) > 0:
        _reasoning_detected = True
    elif getattr(_usage, "reasoning", None) is not None:
        _reasoning_detected = True

if getattr(_message, "reasoning_content", None) is not None:
    _reasoning_detected = True
elif getattr(_message, "reasoning", None) is not None:
    _reasoning_detected = True

if not _reasoning_detected:
    print("Response: ", response)
    raise Exception("VALIDATION FAILED: reasoning - no reasoning information in response")
print("VALIDATION: reasoning SUCCESS")