# OutilsIA Local Cockpit (outilsia.fr)

Scans a PC's GPU, RAM and VRAM, installs a compatible Ollama model with consent, and measures its real tokens-per-second speed.

- Atlas card: https://inite.ai/en/atlas/outilsia.fr
- Site: https://outilsia.fr
- For: people running local AI models on their own Windows or Linux PC, choosing hardware and models
- Categories: llmops
- Use 0/100 · alive 100/100

## Used for

- measure real tokens per second of a local model
- check which local AI model a PC can run
- install Ollama and a compatible model with consent

## Price

- from 39 EUR/one-time (read 2026-10-26 on https://outilsia.fr/pricing)

## For developers

- OpenAPI: https://outilsia.fr/openapi.json (245 paths)
- MCP server: fr.outilsia/local-ai-advisor (namespace; open on 2026-10-26, 15 tools)
- Read 2026-10-26

## MCP server, measured daily

Server: https://outilsia.fr/mcp, checked every day since 2026-10-26 (1 checks).

- Answered: measuring, 1 daily checks so far
- Median first answer: 366 ms
- Sign-in: none
- Context its tool definitions take: about 3216 tokens (15 tools)
- Tool definitions: B, 84/100
- Failing "params_described": 14 of 15 tools
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Read 2026-10-26

---

Source: INITE Atlas, https://inite.ai/en/atlas/outilsia.fr. JSON: https://inite.ai/api/atlas/outilsia.fr. MCP: https://inite.ai/api/mcp (atlas_lookup).