Will this LLM fit your GPU or Mac? npx fitllm — accurate memory math for MLA/sliding-window/hybrid/MoE architectures that naive VRAM calculators get wrong by up to 18x. Single file, zero deps, conformance-vector tested. MIT.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add --transport http fitllm https://fitllm.run/api/mcp~/.codex/config.toml
[mcp_servers.fitllm]
url = "https://fitllm.run/api/mcp"
# codex mcp add fitllm --url https://fitllm.run/api/mcp~/.cursor/mcp.json
{
"mcpServers": {
"fitllm": {
"url": "https://fitllm.run/api/mcp"
}
}
}.vscode/mcp.json
{
"servers": {
"fitllm": {
"type": "http",
"url": "https://fitllm.run/api/mcp"
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"fitllm": {
"httpUrl": "https://fitllm.run/api/mcp"
}
}
}claude_desktop_config.json · Customize › Connectors › Add custom connector
{
"mcpServers": {
"fitllm": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://fitllm.run/api/mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"fitllm": {
"url": "https://fitllm.run/api/mcp"
}
}
}