feat(mistral-ai): update model YAMLs [bot]

[models-bot]

Auto-generated by poc-agent for provider mistral-ai.

---

Note

Low Risk
Metadata-only YAML changes for pricing and feature flags; no runtime or auth logic.

Overview
Updates ministral-14b-latest provider metadata to match other Mistral chat models.

Costs: Adds input_cost_per_token_batches and output_cost_per_token_batches at 1e-7 each (half of the on-demand 2e-7 rates).

Features: Declares system_messages and assistant_prefill so routing and capability checks reflect supported chat patterns.

^{Reviewed by Cursor Bugbot for commit f3226ff. Bugbot is set up for automated code reviews on this repo. Configure here.}

[github-actions]

/test-models

[harshiv-26]

Gateway test results

• Total: 10
• Passed: 8
• Failed: 0
• Validation failed: 2
• Errored: 0
• Skipped: 0
• Success rate: 80.0%

Provider	Model	Scenarios
mistral-ai	ministral-14b-latest	success: tool-call, parallel-tool-call:stream, params, tool-call:stream, params:stream, parallel-tool-call, json-output, json-output:stream validation_failure: structured-output, structured-output:stream

Failures (2)

mistral-ai/ministral-14b-latest — structured-output (validation_failure)

Error

Traceback (most recent call last):
  File "/tmp/tmp01h0bfbn/snippet.py", line 43, in <module>
    raise Exception("VALIDATION FAILED: structured-output - missing expected fields (name, date, participants)")
Exception: VALIDATION FAILED: structured-output - missing expected fields (name, date, participants)

Code snippet

from openai import OpenAI
import json

client = OpenAI(api_key="***", base_url="https://internal.devtest.truefoundry.tech/api/llm")

response_schema = json.loads('''{
  "title": "CalendarEvent",
  "type": "object",
  "properties": {
    "name": { "type": "string" },
    "date": { "type": "string" },
    "participants": {
      "type": "array",
      "items": { "type": "string" }
    }
  },
  "required": ["name", "date", "participants"],
  "additionalProperties": false
}''')

response = client.chat.completions.create(
    model="test-v2-mistral-ai/ministral-14b-latest",
    messages=[
        {"role": "system", "content": "Extract the event information as JSON."},
        {"role": "user", "content": "Hi"},
        {"role": "assistant", "content": "Hi, how can I help you"},
        {"role": "user", "content": "Alice and Bob are going to a science fair on Friday. Extract the event details as JSON."},
    ],
    response_format={"type": "json_schema", "json_schema": {"name": "CalendarEvent", "schema": response_schema}},
    stream=False,
)
import json as _json

_content = response.choices[0].message.content
print(_content)

if not _content:
    raise Exception("VALIDATION FAILED: structured-output - response content is empty")

_parsed = _json.loads(_content)

if "name" not in _parsed or "date" not in _parsed or "participants" not in _parsed:
    raise Exception("VALIDATION FAILED: structured-output - missing expected fields (name, date, participants)")

if not isinstance(_parsed.get("participants"), list):
    raise Exception("VALIDATION FAILED: structured-output - 'participants' is not a list, schema not enforced")

if set(_parsed.keys()) != {"name", "date", "participants"}:
    raise Exception(
        f"VALIDATION FAILED: structured-output - unexpected keys present: {set(_parsed.keys())}"
    )

print("VALIDATION: structured-output SUCCESS")

Output

{
  "title": "CalendarEvent",
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": true
}

mistral-ai/ministral-14b-latest — structured-output:stream (validation_failure)

Error

Traceback (most recent call last):
  File "/tmp/tmpi6ey18de/snippet.py", line 48, in <module>
    raise Exception("VALIDATION FAILED: structured-output stream - missing expected fields (name, date, participants)")
Exception: VALIDATION FAILED: structured-output stream - missing expected fields (name, date, participants)

Code snippet

from openai import OpenAI
import json

client = OpenAI(api_key="***", base_url="https://internal.devtest.truefoundry.tech/api/llm")

response_schema = json.loads('''{
  "title": "CalendarEvent",
  "type": "object",
  "properties": {
    "name": { "type": "string" },
    "date": { "type": "string" },
    "participants": {
      "type": "array",
      "items": { "type": "string" }
    }
  },
  "required": ["name", "date", "participants"],
  "additionalProperties": false
}''')

response = client.chat.completions.create(
    model="test-v2-mistral-ai/ministral-14b-latest",
    messages=[
        {"role": "system", "content": "Extract the event information as JSON."},
        {"role": "user", "content": "Hi"},
        {"role": "assistant", "content": "Hi, how can I help you"},
        {"role": "user", "content": "Alice and Bob are going to a science fair on Friday. Extract the event details as JSON."},
    ],
    response_format={"type": "json_schema", "json_schema": {"name": "CalendarEvent", "schema": response_schema}},
    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: structured-output stream - no content received")

_parsed = _json.loads(_accumulated)

if "name" not in _parsed or "date" not in _parsed or "participants" not in _parsed:
    raise Exception("VALIDATION FAILED: structured-output stream - missing expected fields (name, date, participants)")

if not isinstance(_parsed.get("participants"), list):
    raise Exception("VALIDATION FAILED: structured-output stream - 'participants' is not a list, schema not enforced")

if set(_parsed.keys()) != {"name", "date", "participants"}:
    raise Exception(
        f"VALIDATION FAILED: structured-output stream - unexpected keys present: {set(_parsed.keys())}"
    )

print("\nVALIDATION: structured-output stream SUCCESS")

Output

{
  "title": "CalendarEvent",
  "type": "object",
  "properties": {},
  "required": [],
  "additionalProperties": true,
  "instance": {
    "name": "Science Fair",
    "date": "Friday",
    "participants": ["Alice", "Bob"]
  }
}

Successes (8)

mistral-ai/ministral-14b-latest — tool-call (success)

Output

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

mistral-ai/ministral-14b-latest — parallel-tool-call:stream (success)

Output

{"location": "London"}{"location": "Paris"}
Number of parallel tool calls: 2
VALIDATION: parallel-tool-call stream SUCCESS

mistral-ai/ministral-14b-latest — params (success)

Output

The capital of France is **Paris**! 🇫🇷

Would you like to know more about it? 😊

mistral-ai/ministral-14b-latest — tool-call:stream (success)

Output

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

mistral-ai/ministral-14b-latest — params:stream (success)

Output

The capital of France is **Paris**! 🇫🇷

Would you like to know more about it? 😊

mistral-ai/ministral-14b-latest — parallel-tool-call (success)

Output

Number of parallel tool calls: 2
Function: get_weather
Arguments: {"location": "London"}
Function: get_weather
Arguments: {"location": "Paris"}
VALIDA
... (truncated, 33 chars omitted)

mistral-ai/ministral-14b-latest — json-output (success)

Output

{
  "colors": [
    {
      "name": "Deep Sky Blue",
      "hex_code": "#00BFFF"
    },
    {
      "name": "Forest Green",
      "hex_code": "#228B22
... (truncated, 113 chars omitted)

mistral-ai/ministral-14b-latest — json-output:stream (success)

Output

{
  "colors": [
    {
      "name": "Red",
      "hex_code": "#FF0000"
    },
    {
      "name": "Blue",
      "hex_code": "#0000FF"
    },
    {
   
... (truncated, 99 chars omitted)