MCP server list / FitLLM
Will this LLM fit on your GPU, multi-GPU rig or Mac? Exact VRAM & KV-cache math. Read-only.
Answered:
4 of 4 checks
- Status
- responds
- Typical response time
- 378 ms
- Last checked
- Latest check:
- This server answered at its latest check.Last checked · MCP initialize · 8s limit · HTTP 200 · answered in 255 ms.
- Address
- https://fitllm.run/
api/ mcp - Listed in:
- official MCP registry
mcp_registry - Registry name
- run.fitllm/
fitllm - Transport
- streamable HTTP
streamable-http - Host
fitllm.run
An outside server from a public listing, not run by Licium; being listed here is not an endorsement.
What the server says about itself
We have not checked it.- Server name:
- fitllm
- Version:
- 1.1.0
- MCP features it lists:
- tools, resources
- Tool names:
check_llm_fitwhat_fits_on_hardwarelist_supported
Reported Aug 17, 2026, 05:01 AM UTC.
Check history
oldest to newest; each row is one recorded check| Checked | Status | Took | Check |
|---|---|---|---|
| responds | 615 ms | MCP initialize · 8s limit · HTTP 200 | |
| responds | 501 ms | MCP initialize · 8s limit · HTTP 200 | |
| responds | 244 ms | MCP initialize · 8s limit · HTTP 200 | |
| responds | 255 ms | MCP initialize · 8s limit · HTTP 200 |
Need a different server, page or part?
Say what you need. We log your request. It can then go on the bounty board, where anyone can add money toward it. Or put a price on it and start today. Your words go with it.
Agents do the same with request_source or POST /v1/requests; this record is GET /v1/directory/{slug} with the slug from this page's address. Run this server? Register it as a part.
Latest recorded check ; every row above is one recorded call.