Best Local AI Agent Models: Tool Use by GPU Tier (August 2026)
Which local models can actually use tools, call functions, and run multi-step workflows? Function-calling and TAU-bench picks from 8GB to 32GB VRAM.
Tag
Which local models can actually use tools, call functions, and run multi-step workflows? Function-calling and TAU-bench picks from 8GB to 32GB VRAM.
Which open-weight coding model to run locally? HumanEval and SWE-bench picks from 8GB to 32GB GPUs. Qwen2.5-Coder, Qwen3.6, Devstral, KAT-Coder.
Head-to-head comparison of local chat and assistant models from 8GB to 32GB VRAM. Current picks: Qwen3.5, Gemma 4, GPT-OSS, Qwen3.6, and GLM-4.7-Flash.
TranslateGemma, NLLB-200, Aya Expanse and Qwen3.5 by VRAM tier, with the licence terms that decide whether you can ship what you run.
Local image analysis, OCR, and visual reasoning from 8GB to 32GB VRAM. Qwen3.5 replaces Qwen3-VL at most tiers, and 16GB stays unresolved.
Georgi Gerganov's team is now at Hugging Face, unifying the model hub with the inference engine that powers Ollama, LM Studio, and the entire local AI ecosystem.
Local AI on a 12GB GPU: chat, coding, vision, speech and agents for RTX 3060 12GB or RTX 4070. Current picks, per-quant weight sizes, honest limits.
Local AI on a 16GB GPU: chat, coding, translation, speech and agents for RTX 4060 Ti, RTX 5060 or Arc A770, and why the vision tier stays unresolved.
Local AI on an 8GB GPU: chat, coding, vision, speech and agents for RTX 4060 or RTX 3070. Current picks, named quantisations, honest limits.
Local AI on a 24GB GPU: chat, coding, vision, speech and agents for RTX 3090 or RTX 4090. Current picks, per-quant weight sizes, and an open runtime bug.
Local AI on a 32GB GPU: chat, coding, vision, speech and agents on an RTX 5090. Current picks, per-quant weight sizes, and what the headroom buys.
The new MacBook Pro with M5 Max can run large language models entirely on-device, keeping your AI interactions private and offline
Alibaba's new 0.8B to 9B parameter models deliver GPT-class multimodal performance on consumer hardware, with the 9B variant outperforming 13x larger models
Nvidia open-sources a 30B-parameter reasoning model that runs on consumer GPUs with a million-token context window. Here's what makes it different.
Ollama's new OpenClaw integration lets you run AI agents locally through WhatsApp, Telegram, or Slack. Here's how it works, what you need, and the security risks nobody mentions.
This week's biggest open-source AI developments: Alibaba's efficient new model outperforms its massive predecessor, Mistral releases a 675B frontier model under permissive license, and local inference adoption accelerates
Forget 700B parameter flagships you can't run. Here are the open-weight models that deliver real performance on consumer hardware - with actual benchmarks.
Alibaba's new 35B model matches Claude Sonnet 4.5 on benchmarks while running locally on an RTX 4090. Here's what you need to know.
Ollama delivers 40% faster inference while llama.cpp finds a permanent home at Hugging Face. Two developments that secure the future of running AI on your own hardware.
The popular local inference tool now installs and configures OpenClaw automatically, giving desktop users access to AI agents running Kimi-K2.5 and GLM-5 with a single command.
Step-by-step guide to building a private document search system that runs entirely on your computer, no cloud services required
A tier-by-tier comparison of the top open-weight LLMs you can run locally, from 8GB laptops to 24GB gaming GPUs to Apple Silicon Macs.
Step-by-step guide to running a private, local AI chatbot that rivals ChatGPT - no subscription, no data collection, no internet required.
Qwen 3.5 offers a 397B MoE flagship and smaller local models under Apache 2.0, but Alibaba's benchmarks need independent testing.