feat(models): add DeepSeek model configurations

[JasonOA888]

Fixes #215

Add configurations for DeepSeek-V3 and DeepSeek-Chat models via OpenRouter integration.

Models Added

• DeepSeek-V3: Reasoning model (input: $0.27/1M, output: $1.10/1M)
• DeepSeek-Chat: Chat model (input: $0.14/1M, output: $0.28/1M)

Usage

Set OPENROUTER_API_KEY in .env and use:

python eval/swe_bench/evaluate.py --model deepseek-v3

Alternative: Use native DeepSeek API with api_key and base_url config.

[abhigyanpatwari]

🟢 GitNexus Blast Radius: LOW

Metric	Count
Changed symbols	0
Direct dependents (d=1)	0
Indirect (d=2)	0
Transitive (d=3)	0
Flows impacted	0
Total affected	0

Changed: None detected
Flows hit: None

View full blast radius graph →

---

Generated by GitNexus — code intelligence powered by knowledge graphs

[vercel]

@JasonOA888 is attempting to deploy a commit to the NexusCore Team on Vercel.

A member of the Team first needs to authorize it.

[reversTeam]

Looks good, @JasonOA888! Clean config-only addition for DeepSeek models.

A couple of minor notes:

• The commented-out OpenRouter config blocks are a nice touch for discoverability.
• API keys are properly set to null (not hardcoded), which is correct.
• Cost annotations per 1M tokens are helpful for eval budgeting.

LGTM — straightforward config addition with no code risk.