Can this LLM run on my GPU? VRAM, speed ceiling and what fits instead, for any model and GPU.
Scored across 6 tools
The tools split cleanly along direction (model→GPU via can_i_run, GPU→model via what_fits) and along resource listing (list_gpus, list_models), but estimate_vram and estimate_from_hf_repo overlap heavily since estimate_vram already accepts 'any Hugging Face repo id', and can_i_run repeats estimate_vram's memory-split output. An agent can pick a reasonable tool, but the boundary between the three estimation tools requires reading descriptions carefully.
All names are snake_case and readable, with a dominant verb_noun or verb_preposition pattern (list_models, estimate_vram, estimate_from_hf_repo). The two question-phrased names (can_i_run, what_fits) break the pattern slightly, though they usefully signal the user-intent tools.
Six tools is well-scoped for a VRAM-estimation server: two catalogues, three estimation paths, one reverse lookup. Every tool has a clear job and none feels like filler.
The surface covers the full question lifecycle: enumerate known GPUs and models, estimate memory for a known or arbitrary HF model, check a specific model/GPU pair, and reverse-search what fits a card. Minor gaps: no batch/multi-GPU comparison and no way to pass runtime/tokenizer or speed constraints beyond the built-in decode ceiling.
Can this GPU run this open-weight LLM? Returns fits, tight or no, the memory split (weights, KV cache, overhead), a decode-speed ceiling, the longest context that fits and, on a no, every change that would make it fit: quantisation, KV cache, context, another card or a smaller model. Model: a name or id from list_models, any Hugging Face repo id, or its architecture. GPU: a name or id from list_gpus, or vram_gb for any other card.
Inputs: gpu, model, quant, context, +6 more.
Reads any Hugging Face model repo's config.json and parameter count and estimates its memory: the attention layout found (standard, sliding-window, hybrid or latent), how much each 1,000 tokens of context costs, and weights + KV cache + overhead at every quantisation. For models nodegrove.io has not reviewed; anything the reader cannot model is listed in warnings.
Inputs: repo (required), context, kv_cache, active_params_b.
How much memory an LLM needs: weights + KV cache + overhead at each quantisation (or one), at a given context, and the smallest common card class that holds each. Model: a name or id from list_models, any Hugging Face repo id, or its architecture (params_b, layers, kv_heads, head_dim).
Inputs: model, quant, context, kv_cache, +2 more.
The GPUs and machines nodegrove.io covers: memory, the memory a runtime can use and bandwidth, from the makers' specs, with each one's page.
Inputs: search.
The open-weight LLMs nodegrove.io has verified against their config.json (data version 2026-11-03): id, size, attention design, native context, licence, memory at Q4 with 8k context and each model's page.
Inputs: search.
Which open-weight LLMs fit this GPU: every model in list_models checked at one quantisation and context, with a recommended everyday model (the biggest class that fits with room for context at conversational speed), the largest that fits, the best at Q8 and the first out of reach. GPU: a name or id from list_gpus, or vram_gb for any other card.
Inputs: gpu, quant, context, vram_gb, +2 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Does an LLM fit on your GPU? VRAM, KV cache and GPU count for any Hugging Face model.
Will this LLM fit on your GPU, multi-GPU rig or Mac? Exact VRAM & KV-cache math. Read-only.
Will a local LLM run on your hardware? GGUF quant, buy-vs-rent-vs-API cost, used-GPU prices.
Read-only: which local LLMs a GPU or Mac can run - VRAM fit, tokens/sec, model specs.
Add one secure layer between your agents and this server.
