# RigBase (getrigbase.com)

Publishes measured tokens-per-second benchmarks of local LLMs across home GPU and unified-memory rigs, each result linked to its source.

- Atlas card: https://inite.ai/en/atlas/getrigbase.com
- Site: https://getrigbase.com
- For: People choosing a GPU, Mac or home AI rig to run local LLMs such as Qwen3 or Gemma
- Categories: llmops
- Use not measured/100 · alive 100/100

## Used for

- check whether a model fits your GPU's VRAM
- compare measured tokens per second across home AI rigs
- look up benchmark results for a specific model and GPU

## Price

Not read yet.

## For developers

- MCP server: com.getrigbase/rigbase (namespace; open on 2026-11-01, 4 tools)
- Read 2026-11-01

## MCP server, measured daily

Server: https://getrigbase.com/mcp, checked every day since 2026-11-01 (2 checks).

- Answered: measuring, 2 daily checks so far
- Median first answer: 151 ms
- Sign-in: none
- Context its tool definitions take: about 650 tokens (4 tools)
- Tool definitions: A, 100/100
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Read 2026-11-01

---

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