The request names GLM-4.5-Air-GGUF, which none of the candidate sources describe directly. The most relevant products for running GGUF model files locally are Hugging Face, whose local-models and hf-cli actions cover finding and downloading GGUFs 212, and Ollama, which runs open models locally on macOS, Windows, and Linux 5, with AI Compute Radar offering GGUF-based fit checks for your GPU 8.
A question back: Do you want to run GLM-4.5-Air-GGUF on your own hardware (which GPU or Mac, and how much memory), or access it through a hosted API?
- 1
Hugging Facehuggingface.coverified and signed
Hosts public models and datasets, and its hf-cli action downloads and manages models from the Hub, while huggingface-local-models covers finding GGUFs and selecting quantizations to run locally with llama.cpp 212.
official:MCP server · @huggingface/mcp-clientskill · hf-cliskill · hf-cloud-aws-context-discoveryskill · hf-cloud-python-env-setupskill · hf-cloud-sagemaker-deployment-planner
for agents: huggingface-local-models · hf-mem · hf-cli
- 2
Ollamaollama.comwebsite read, tools not yet checked
Runs open AI models locally or in the cloud on macOS, Windows, and Linux, giving a simple way to run GGUF-style open models on your own machine 5.
+137 community tools built for it
- 3
AI Compute Radaraicomputeradar.devwebsite read, tools not yet checked
Ranks open AI models and checks whether a model fits your GPU or Mac using real GGUF file sizes 8.
official:MCP server · AI Compute Radar
Sources
- 2Hugging Face – The AI community building the future.huggingface.co · read 2026-10-29
- 5Ollama · Run AI models locally and in the cloudollama.com · read 2026-10-29
- 8AI Compute Radar — open models ranked, matched to your GPUaicomputeradar.dev · read 2026-10-29
- 12huggingface-local-models: declares "huggingface-local-models" (the record we read and signed)unfragile.ai · read 2026-10-24
For agents: the 10 actions behind this answer
spuuntries/flatdolphinmaid-8x7b-gguf · spuuntries/flatdolphinmaid-8x7b-gguf verified and signed
Undi95's FlatDolphinMaid 8x7B Mixtral Merge, GGUF Q5_K_M quantized by TheBloke.
spuuntries/miqumaid-v1-70b-gguf · spuuntries/miqumaid-v1-70b-gguf verified and signed
NeverSleep's MiquMaid v1 70B Miqu Finetune, GGUF Q3_K_M quantized by NeverSleep.
huggingface-local-models · huggingface-local-models verified and signed
Use to select models to run locally with llama.cpp and GGUF on CPU, Mac Metal, CUDA, or ROCm. Covers finding GGUFs, quant selection, running servers, exact GGUF file lookup, conversion, and OpenAI-compatible local serving.
cuuupid/glm-4v-9b · cuuupid/glm-4v-9b verified and signed
GLM-4V is a multimodal model released by Tsinghua University that is competitive with GPT-4o and establishes a new SOTA on several benchmarks, including OCR.
workers-ai-download-finetune-asset · Cloudflare API verified and signed
Download a Finetune Asset
arize-trace · arize-trace verified and signed
Downloads, exports, and inspects existing Arize traces and spans to understand what an LLM app is doing or debug runtime issues. Covers exporting traces by ID, spans by ID, sessions by ID, and root-cause investigation using the ax CLI. Use when the user wants to look at existing trace data, see what
· Ollama website read, tools not yet checked
Ollama is the easiest way to run and build with open AI models locally or in the cloud on macOS, Windows, and Linux.
github · ghapi verified and signed
kcaverly/deepseek-coder-33b-instruct-gguf · kcaverly/deepseek-coder-33b-instruct-gguf verified and signed
A quantized 33B parameter language model from Deepseek for SOTA repository level code completion
bytedance/seedream-4.5 · bytedance/seedream-4.5 verified and signed
Seedream 4.5: Upgraded Bytedance image model with stronger spatial understanding and world knowledge
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.