Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Official-source-led manual for the TUI `nvidia` template.
.claude/skills/lingtai-ai-preset-skill-nvidia/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -61% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -71% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -77% | 0% |
| case-14 | ✗→✓ | ▲ Improved | -68% | 0% |
nvidianvidiaPreset() in tui/internal/preset/preset.go ships exact model meta/llama-3.3-70b-instruct at https://integrate.api.nvidia.com/v1 with OpenAI compatibility and NVIDIA_API_KEY. This default model is text-only and the manifest has no vision capability.
The same hosted gateway may offer other VLM slugs, but a clone selecting one still needs an explicit capabilities.vision entry; changing only llm.model does not wire a LingTai vision route. Read the official NVIDIA API Catalog, model catalog, and NIM API reference on demand. No NVIDIA plan-level vision MCP is established. Do not switch providers or auto-load/invoke an MCP. Recheck TUI source for the template fields after changes. Never guess a key or copy the volatile catalog into this manual.
For base URL/API-compat/model/capability declaration shape versus credentials, see reference/operations/endpoint-capabilities/SKILL.md.
Other measured skills in the registry, with their headline benchmark lift.