fix: add Together models deepseek-ai/DeepSeek-V4-Pro and Qwen/Qwen3.5-397B-A17B

[github-actions]

fix: add Together models deepseek-ai/DeepSeek-V4-Pro and Qwen/Qwen3.5-397B-A17B

Closes #515

Source issue: #515

Summary

Field	Value
Provider	together
Primary model	deepseek-ai/DeepSeek-V4-Pro
Changed models	deepseek-ai/DeepSeek-V4-Pro Qwen/Qwen3.5-397B-A17B
Added models	deepseek-ai/DeepSeek-V4-Pro Qwen/Qwen3.5-397B-A17B
Updated models	None
Verification sources	1 2 3 4

Verified metadata

Model	Display name	Parent	Providers	Format	Flavor	Token limits	Pricing	Lifecycle
deepseek-ai/DeepSeek-V4-Pro	DeepSeek V4 Pro		together	openai	chat	input=512000, output=not provided	in/out=2.1/4.4 per 1M; cache read=0.2 per 1M	active
Qwen/Qwen3.5-397B-A17B	Qwen3.5 397B A17B		together	openai	chat	input=262144, output=not provided	in/out=0.6/3.6 per 1M; cache read=0.06 per 1M	active

Verification notes

No LLM verification step ran — model metadata was already complete in the issue.

sync_models vs proposed update

sync_models cross-check found differences. Official provider verification was used for the applied values, and sync_models discrepancies are listed below for review.

Model	Field	Proposed update	sync_models	sync_models source models
deepseek-ai/DeepSeek-V4-Pro	catalog entry	present	missing	None
Qwen/Qwen3.5-397B-A17B	input_cache_read_cost_per_mil_tokens	0.06	n/a	together_ai/Qwen/Qwen3.5-397B-A17B

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
ai-proxy	Ready	Preview, Comment	May 4, 2026 8:40pm

[github-actions]

Proxy model verification

Verification completed successfully.

Passed: 2
Failed: 0

Tested models

• deepseek-ai/DeepSeek-V4-Pro
• Qwen/Qwen3.5-397B-A17B

Passed models

• deepseek-ai/DeepSeek-V4-Pro
• Qwen/Qwen3.5-397B-A17B

Failed models

• None