[BOT ISSUE] Together: add missing `together` provider to deepseek-ai/DeepSeek-V4.1-Flash

[github-actions]

Summary

The model deepseek-ai/DeepSeek-V4.1-Flash is listed on Together AI's serverless inference surface but together is missing from its available_providers in model_list.json. The entry currently lists only ["baseten", "fireworks"].

Gap details

Field	Current (local)	Together AI official
available_providers	["baseten", "fireworks"]	Should include "together"
input_cost_per_mil_tokens	0.3	0.30 ✓ (matches)
output_cost_per_mil_tokens	1.2	1.20 ✓ (matches)
input_cache_read_cost_per_mil_tokens	0.03	0.006 (Together's rate; local value reflects baseten rate)

Base input and output pricing matches exactly between the local catalog and Together's listing. The cached input price differs ($0.03 locally from baseten vs $0.006 on Together) — since the catalog uses a single price field, the existing value can remain.

Verification checklist

Check	Status	Detail
Cross-source	✅	Model confirmed on Together serverless models page (model listing + $0.30/$1.20 pricing) AND Together pricing page ($0.30/$1.20/$0.006 cached). Model also already in catalog via baseten and fireworks
Already in catalog	✅	deepseek-ai/DeepSeek-V4.1-Flash exists with available_providers: ["baseten", "fireworks"] — together is absent
ID format	✅	Together uses the same deepseek-ai/DeepSeek-V4.1-Flash model ID, matching the existing catalog entry
Duplicate check	✅	No existing open or closed issue covers adding together to this model's providers

Verification notes

Field	Value	Source
Model ID deepseek-ai/DeepSeek-V4.1-Flash on Together	Verified	Together serverless models page — listed in Chat models table
Input $0.30/MTok	Verified	Together pricing page
Output $1.20/MTok	Verified	Together pricing page
Cached input $0.006/MTok	Verified (Together-specific rate)	Together pricing page
Supports vision (multimodal)	✅	Listed under both Chat and Vision tabs on Together serverless models page

Local files inspected

• packages/proxy/schema/model_list.json — deepseek-ai/DeepSeek-V4.1-Flash at line ~11254, available_providers: ["baseten", "fireworks"]

{
  "kind": "stale_metadata",
  "provider": "together",
  "models": ["deepseek-ai/DeepSeek-V4.1-Flash"],
  "status": "active",
  "model_specs": {
    "deepseek-ai/DeepSeek-V4.1-Flash": {
      "format": "openai",
      "flavor": "chat",
      "multimodal": true,
      "input_cost_per_mil_tokens": 0.3,
      "output_cost_per_mil_tokens": 1.2,
      "input_cache_read_cost_per_mil_tokens": 0.03,
      "displayName": "DeepSeek V4.1 Flash",
      "reasoning": true,
      "max_input_tokens": 1048576,
      "available_providers": ["baseten", "fireworks", "together"]
    }
  },
  "source_urls": [
    "https://docs.together.ai/docs/serverless-models",
    "https://www.together.ai/pricing"
  ]
}

[github-actions]

Prepared catalog updates for deepseek-ai/DeepSeek-V4.1-Flash. Updated models: deepseek-ai/DeepSeek-V4.1-Flash.

Verification sources: https://docs.together.ai/docs/serverless-models, https://www.together.ai/pricing
Verified metadata applied:

• deepseek-ai/DeepSeek-V4.1-Flash: format=openai, flavor=chat, displayName=DeepSeek V4.1 Flash, max_input_tokens=1048576, max_output_tokens=not provided, pricing(input/output per 1M)=0.3/1.2, input_cache_read_cost_per_mil_tokens=0.03, providers=baseten|fireworks|together, reasoning=true, multimodal=true

Included in the daily batch PR: #1247