mcpcost

org.marvinlabs/vram-fit

MarvinLabs vram_fit

Is the MarvinLabs vram_fit MCP server down? No. It answered initialize and tools/list in the probe of 2026-10-26.

Will a QLoRA fine-tune of gemma-3-4b-it fit your GPU? Verdicts from measured AMD and NVIDIA probes.

alive probed 2026-11-15 · unchanged since 2026-10-26 · HTTP 200

3tools
1.1ktokens in Claude Opus 5.5
0.6 %of Claude's 200k window
0.4×the median server
1.1 sto initialize
817 msto list tools
Endpoint
https://marvinlabs.org/mcp (streamable-http)
Host
marvinlabs.org
Protocol
2025-06-18
Reports itself as
vram_fit 0.1.0
Capabilities
tools
Instructions
287 characters
Of a 1M window
0.1 % of a 1M context window
Count
1.1k tokens in Claude Opus 5.5, exact, measured 2026-10-26. With the Claude Code prefix mcp__…__ on each tool name: 1.1k. The o200k_base estimate is 659. Counted again when the tool definitions change. Method.
Website
https://marvinlabs.org/vram
Registry
active, updated 2026-10-25 · registry entry

Most expensive tools

The Claude column scales the o200k estimate of each tool by this server's ratio of exact to estimate, 1.68×.

Tool≈ Claude tokenso200k estimateShare of this server
vram_fit49329344 %
vram_max_rank42625338 %
vram_probes19011317 %

Compare this server with others.

More servers

Servers that mention vram.

Badge for the README

mcpcost: alive

The result of the latest probe, the tool count and the token cost. The feed has an entry for every probe that changed them.

Atom feed

A mail when this server changes

Stops answering, comes back, tool set or cost changes. One confirmation mail first, a stop link in every mail.

For the publisher

Fixed something? Probe it now, with the same steps as the daily run. This page updates with the next daily run.