[BOT ISSUE] Mistral: add missing reasoning flags for Medium 3.5 and Small 4

[github-actions]

Summary

Mistral Medium 3.5 and Mistral Small 4 both support the reasoning_effort parameter (with values "high" and "none"), but their catalog entries lack reasoning: true and reasoning_budget: true flags. This means the proxy does not advertise these models as reasoning-capable.

Mistral's deprecated Magistral models were native reasoning models; the current Medium 3.5 and Small 4 replaced them with adjustable reasoning via the reasoning_effort parameter instead.

Affected Entries

Model ID	Current reasoning	Correct Value	Line
mistral-medium-latest	(missing)	true	~4357
mistral-medium-3.5	(missing)	true	~4772
mistral-small-latest	(missing)	true	~4382

Also affected (snapshots and aliases with the same gap):

• mistral-medium-3 (line ~4784) — parent: mistral-medium-3.5
• mistral-small-2603 (line ~4746) — parent: mistral-small-latest

All other fields (pricing, format, context window) are correct and unchanged.

Verification

Check	Status	Detail
Cross-source (reasoning support)	✅	Mistral reasoning docs: explicitly lists mistral-small-latest and mistral-medium-3-5 as supporting reasoning_effort AND Mistral models overview: lists both as current active models with reasoning capability
Already fixed	✅	Confirmed all 3 primary entries lack reasoning and reasoning_budget fields
Sibling consistency	✅	Other reasoning models in the catalog (gpt-5.4-mini, claude-fable-5, claude-opus-4-8) all have reasoning: true
Duplicate check	✅	No open issue covers Mistral reasoning flags

Verification Notes

Field	Source	Notes
reasoning_effort support	Mistral reasoning docs	"mistral-small-latest" and "mistral-medium-3-5" both support reasoning_effort with values "high" and "none"
Magistral deprecation	Mistral reasoning docs	Native reasoning models (magistral) deprecated; replaced by reasoning_effort on Small 4 and Medium 3.5
reasoning_budget: true	Catalog convention	Models supporting adjustable reasoning effort get reasoning_budget: true (e.g., claude-fable-5, claude-opus-4-8)

Local Files Inspected

• packages/proxy/schema/model_list.json — all 3 primary entries confirmed to lack reasoning and reasoning_budget fields

{
  "kind": "stale_metadata",
  "provider": "mistral",
  "models": ["mistral-medium-latest", "mistral-medium-3.5", "mistral-small-latest"],
  "status": "active",
  "model_specs": {
    "mistral-medium-latest": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 1.5,
      "output_cost_per_mil_tokens": 7.5,
      "displayName": "Mistral Medium 3.5",
      "reasoning": true,
      "reasoning_budget": true,
      "max_input_tokens": 256000,
      "available_providers": ["mistral"]
    },
    "mistral-medium-3.5": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 1.5,
      "output_cost_per_mil_tokens": 7.5,
      "displayName": "Mistral Medium 3.5",
      "reasoning": true,
      "reasoning_budget": true,
      "max_input_tokens": 256000,
      "available_providers": ["mistral"]
    },
    "mistral-small-latest": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 0.15,
      "output_cost_per_mil_tokens": 0.6,
      "displayName": "Mistral Small",
      "reasoning": true,
      "reasoning_budget": true,
      "max_input_tokens": 262144,
      "max_output_tokens": 131072,
      "available_providers": ["mistral"]
    }
  },
  "source_urls": [
    "https://docs.mistral.ai/capabilities/reasoning/",
    "https://docs.mistral.ai/getting-started/models"
  ]
}

[github-actions]

Prepared catalog updates for mistral-medium-latest, mistral-medium-3.5, mistral-small-latest. Updated models: mistral-medium-latest, mistral-medium-3.5, mistral-small-latest.

Verification sources: https://docs.mistral.ai/capabilities/reasoning/, https://docs.mistral.ai/getting-started/models
Verified metadata applied:

• mistral-medium-latest: format=openai, flavor=chat, displayName=Mistral Medium 3.5, max_input_tokens=256000, max_output_tokens=not provided, pricing(input/output per 1M)=1.5/7.5, providers=mistral, reasoning=true, reasoning_budget=true, multimodal=true
• mistral-medium-3.5: format=openai, flavor=chat, displayName=Mistral Medium 3.5, max_input_tokens=256000, max_output_tokens=not provided, pricing(input/output per 1M)=1.5/7.5, providers=mistral, reasoning=true, reasoning_budget=true, multimodal=true
• mistral-small-latest: format=openai, flavor=chat, displayName=Mistral Small, max_input_tokens=262144, max_output_tokens=131072, pricing(input/output per 1M)=0.15/0.6, providers=mistral, reasoning=true, reasoning_budget=true, multimodal=true

Included in the daily batch PR: #904