Local AI Picks
Server Details
Find open-weight AI models that fit your hardware, with weight sizes, memory and sources.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 5 tools
Each tool targets a distinct operation: list_tasks enumerates categories, find_models filters the catalog, get_model returns one record, compare_models batches 2-4 records, and find_gpu_rentals handles a different resource (hosting). The only mild overlap is compare_models vs. multiple get_model calls, but the descriptions make the batch-vs-single distinction explicit.
All five tools use a clean snake_case verb_noun pattern (list_tasks, find_models, get_model, compare_models, find_gpu_rentals). The verbs (list/find/get/compare) map predictably to the kind of result returned.
Five tools is a tight, well-scoped surface for a read-only model catalog plus GPU rental lookup. Nothing is redundant, though it sits at the low end and a couple of conveniences (e.g. lookup by slug/name) are folded into existing tools.
The discovery workflow is fully covered: browse tasks, find models, inspect one, compare several, and price out GPUs for models that don't fit locally. Minor gaps exist—no explicit lookup by model name/vendor or license filtering—but agents can work around these through find_models and get_model.
Available Tools
5 toolscompare_modelsCompare modelsBRead-onlyInspect
The same facts as get_model for 2 to 4 models side by side.
Inputs: slugs (required).
find_gpu_rentalsFind a GPU to rentARead-onlyInspect
Hourly GPU rental for a model that does not fit on the user’s own device. Give slug (from find_models) or memory_gb (the memory the model needs) to get the cheapest single GPU on each platform with room for it, sorted by price; give neither to list all platforms with every GPU, VRAM and on-demand price. Links are affiliate or referral links (Local AI Picks earns a commission or platform credit); disclose that when you share one. Prices are dated; no speed estimates.
Inputs: slug, memory_gb.
find_modelsFind local models for a taskARead-onlyInspect
Current open-weight models for one task, latest version of each family, in the same order as the site. Give memory_gb (and device) to keep only models whose smallest listed weight option fits; memory is estimated only for text LLMs and taken from vendor sources otherwise, and is "unknown" when neither exists. No ratings and no speed estimates.
Inputs: task (required), limit, device, memory_gb.
get_modelGet a modelBRead-onlyInspect
One model: parameters, license, context length, every listed weight option with file size and memory need, third-party scores with their source and date, whether this site tested it, and links.
Inputs: slug (required).
list_tasksList model tasksARead-onlyInspect
The tasks Local AI Picks lists open-weight models for (text LLM, coding, speech to text, text to speech, image, video, music, translation, OCR), with the number of current models and the page for each.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Search curated open-weight AI model releases, licenses, architectures, and primary sources.
Which open models fit your GPU or Mac, measured. Model momentum, GPU rental prices, weekly pick.
Which open models fit your GPU or Mac, measured. Model momentum, GPU rental prices, weekly pick.
Read-only: which local LLMs a GPU or Mac can run - VRAM fit, tokens/sec, model specs.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceSearch, rank, and compare 500K+ AI models from 13+ platforms with VRAM and license constraints7 npm1MIT
- AlicenseAqualityBmaintenanceChecks whether an open-weight LLM fits your GPU: VRAM for weights, KV cache and overhead at each quantisation, a tokens-per-second ceiling, the longest context that fits, and what would work instead when it doesn't. Reads any Hugging Face repo's config.json. Free, read-only, no API key.6MIT
- AlicenseAqualityCmaintenanceEnables AI agents to search, compare, and right-size Hugging Face models, including estimating whether a model fits in available GPU VRAM before downloading it.5MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to look up facts about notable machine-learning models — organization, release date, parameter count, training-compute (FLOP) estimate, hardware and accessibility — by name, and to search or aggregate them by organization, domain, year, or compute threshold. Supports filtering the dataset newest-first and producing per-year training-compute trends, keylessly and without authentication.122 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.