DumboDhruvi/ModelFit-MCP
v0.1.1 github Python analyzed 42d7d1a
Hardware-Aware Hugging Face Discovery, Memory Fitting, Empirical Benchmarking & Swappable Local Model Gateway for AI Agents
DumboDhruvi/ModelFit-MCP is an MCP server distributed on github, maintained by DumboDhruvi, tracked here since September 2026. It has shipped 3 releases (currently 0.1.1) and exposes 9 tools. Tools include benchmark_models, ensemble_predict, find_ensemble_models, get_active_model_status, get_hardware_specs, get_integration_code, and 3 more. Its composite risk grade is A — an inferred review prompt computed from observed signals, not a verdict.
tools
-
benchmark_modelsRun live accuracy and latency benchmarking on candidate models. -
ensemble_predictExecute ensemble prediction combining multiple models. -
find_ensemble_modelsFind ultra-fast models suitable for low-latency ensembling. -
get_active_model_statusInspect currently loaded model and target device. -
get_hardware_specsDetect host CPU, RAM, and GPU/VRAM headroom. -
get_integration_codeGet drop-in Python inference code. -
recommend_and_scaffold1-shot model search, hardware check, and code scaffolding. -
search_compatible_modelsSearch Hugging Face models strictly filtered by hardware fit. -
swap_active_modelHot-swap the active model with VRAM-safe memory purging.
release cadence · 90d
2026-07-14 2026-08-27 now
recent releases
| version | date | src |
|---|---|---|
| 0.1.1 | 2026-09-24 | pypi |
| 0.1.0 | 2026-09-23 | pypi |
| v0.1.0 | 2026-09-23 | github |