[BOT ISSUE] Missing Mistral Magistral 1.2 dated versions (magistral-medium-2509, magistral-small-2509)

[github-actions]

Summary

Mistral released Magistral Medium 1.2 and Magistral Small 1.2 (September 2025), which are newer versions of the Magistral reasoning models. The local catalog has the older magistral-medium-2506 and magistral-small-2506 versions, plus the -latest aliases, but is missing the 2509 dated snapshots. The 2509 versions also have a significantly larger context window (128k vs 40k in the current catalog entries).

Gap Details

Missing Model ID	Replaces	Key difference
magistral-medium-2509	magistral-medium-2506	128k context (catalog has 40k), same pricing
magistral-small-2509	magistral-small-2506	128k context (catalog has 40k), same pricing

Verified Fields

magistral-medium-2509

• Pricing: $2.00 / M input tokens, $5.00 / M output tokens (from model card)
• Context window: 128k tokens (from model card — catalog currently says 40k)
• Alias: magistral-medium-latest (should resolve to this version)
• Release date: September 18, 2025

magistral-small-2509

• Pricing: $0.50 / M input tokens, $1.50 / M output tokens (from model card)
• Context window: 128k tokens (from model card — catalog currently says 40k)
• Alias: magistral-small-latest (should resolve to this version)
• Release date: September 18, 2025

Fields NOT verified

• max_output_tokens — not explicitly published on either model card page
• multimodal — described as "multimodal reasoning model" but specific modalities not confirmed
• reasoning / reasoning_budget — not explicitly stated in API terms on the model card

Official Sources

1. Magistral Medium 1.2 model card: https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09 — confirms API ID magistral-medium-2509, pricing ($2/$5), 128k context
2. Magistral Small 1.2 model card: https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09 — confirms API ID magistral-small-2509, pricing ($0.5/$1.5), 128k context
3. Models overview page: https://docs.mistral.ai/models/overview — lists both as current active models

Local Files Inspected

• packages/proxy/schema/model_list.json — grep for magistral-medium-2509 and magistral-small-2509 returns no matches
• Catalog currently has magistral-medium-2506 (line ~4656, 40k context) and magistral-small-2506 (line ~4680, 40k context)
• magistral-medium-latest (line ~4644) and magistral-small-latest (line ~4668) both still have max_input_tokens: 40000

Verification Notes

• Source 1 & 2 (model cards) verify: API IDs, pricing, 128k context
• Source 3 (overview) verifies: models are current/active, not deprecated
• The -latest aliases should be updated to point to these versions and have their max_input_tokens updated from 40000 to 128000

Relationship to Existing Issues

• Issue [BOT ISSUE] Mistral "latest" aliases have stale pricing, context windows, and incorrect API model IDs #486 (closed) mentioned magistral-medium-2509 in context of stale -latest metadata, but that fix did not add the 2509 dated versions to the catalog
• No open issue covers these missing model IDs

{
  "kind": "missing_model",
  "provider": "mistral",
  "models": ["magistral-medium-2509", "magistral-small-2509"],
  "status": "active",
  "model_specs": {
    "magistral-medium-2509": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 2,
      "output_cost_per_mil_tokens": 5,
      "displayName": "Magistral Medium 1.2 (2509)",
      "parent": "magistral-medium-latest",
      "max_input_tokens": 128000,
      "available_providers": ["mistral"]
    },
    "magistral-small-2509": {
      "format": "openai",
      "flavor": "chat",
      "input_cost_per_mil_tokens": 0.5,
      "output_cost_per_mil_tokens": 1.5,
      "displayName": "Magistral Small 1.2 (2509)",
      "parent": "magistral-small-latest",
      "max_input_tokens": 128000,
      "available_providers": ["mistral"]
    }
  },
  "source_urls": [
    "https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09",
    "https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09",
    "https://docs.mistral.ai/models/overview"
  ]
}

[github-actions]

Prepared catalog updates for magistral-medium-2509, magistral-small-2509. Added models: magistral-medium-2509, magistral-small-2509.

Verification sources: https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09, https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09, https://docs.mistral.ai/models/overview
Verified metadata applied:

• magistral-medium-2509: format=openai, flavor=chat, displayName=Magistral Medium 1.2 (2509), parent=magistral-medium-latest, max_input_tokens=128000, max_output_tokens=not provided, pricing(input/output per 1M)=2/5, providers=mistral
• magistral-small-2509: format=openai, flavor=chat, displayName=Magistral Small 1.2 (2509), parent=magistral-small-latest, max_input_tokens=128000, max_output_tokens=not provided, pricing(input/output per 1M)=0.5/1.5, providers=mistral

Opened #508 to apply the catalog update.

[github-actions]

Prepared catalog updates for magistral-medium-2509, magistral-small-2509. Added models: magistral-medium-2509, magistral-small-2509.

Verification sources: https://docs.mistral.ai/models/model-cards/magistral-medium-1-2-25-09, https://docs.mistral.ai/models/model-cards/magistral-small-1-2-25-09, https://docs.mistral.ai/models/overview
Verified metadata applied:

• magistral-medium-2509: format=openai, flavor=chat, displayName=Magistral Medium 1.2 (2509), parent=magistral-medium-latest, max_input_tokens=128000, max_output_tokens=not provided, pricing(input/output per 1M)=2/5, providers=mistral
• magistral-small-2509: format=openai, flavor=chat, displayName=Magistral Small 1.2 (2509), parent=magistral-small-latest, max_input_tokens=128000, max_output_tokens=not provided, pricing(input/output per 1M)=0.5/1.5, providers=mistral

Opened #508 to apply the catalog update.