Custom Qwen3.5 variants optimized for 128GB unified memory systems, such as AMD Ryzen AI Max+ 395. On Windows 11, GPU is limited to 96GB (32GB reserved for OS/CPU), requiring context window capped at 131072 tokens (128K) to fit within GPU memory limits.
368 Pulls 2 Tags Updated 4 months ago
Qwen 3.6 27B (Q4_K_M) optimized for Hermes Agent — 64K context, 8192 max tokens, MTP for speed, flash attention + Q8 KV cache.
27 Pulls 1 Tag Updated 5 days ago
maximum 256k context length by default
1,506 Pulls 1 Tag Updated 9 months ago
maximum 256k context length for coding and other long-context tasks
835 Pulls 1 Tag Updated 9 months ago
DavidAU/Qwen3.6-40B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-NEO-CODE-Di-IMatrix-MAX-GGUF
1,563 Pulls 1 Tag Updated 2 months ago
qwen3-coder with tools calling, context 58k to match full memory 31GB on RTX5090
567 Pulls 1 Tag Updated 11 months ago
Made for a rig with TRX 3080 10Gb, 32gb RAM and Ryzen 5800x
18 Pulls 1 Tag Updated 2 weeks ago
Alice is a roleplay LLM based on qwen3:4b-instruct-2507-q4_K_M, simulating Alice Marie Chen, a 20yo from Portland. Features an 8k context window and is optimized for WhatsApp. The persona is an anti-assistant with a texting style, refusing technical tasks
622 Pulls 1 Tag Updated 5 months ago
Qwen3.6-35B-A3B MoE coding agent for Claude Code / Codex / opencode, 64K context, native tool-calling, honest tool use, safety guardrails intact
2,624 Pulls 2 Tags Updated 1 month ago
92 Pulls 1 Tag Updated 1 year ago
A memory-efficient model configuration of Qwen3.6-35B-A3B using an upstream imatrix-calibrated IQ4_XS quantization and q4_0 KV cache. Designed for 24 GB VRAM
709 Pulls 1 Tag Updated 1 month ago
Qwen3.5-35B-A3B APEX GGUF -- A Novel MoE-Aware Mixed-Precision Quantization Technique Brought to you by the LocalAI team -- the creators of LocalAI the open-source AI engine that runs any model - LLMs, vision, image - on any hardware.
665 Pulls 2 Tags Updated 3 months ago
Fine-tuned manim model using Qwen2.5-Coder-14B-Instruct trained on 3blue1brown-manim dataset with 2,407 examples.
458 Pulls 3 Tags Updated 5 months ago
The most powerful open-source coding AI - 480B parameters with Mixture of Experts architecture for exceptional code generation and understanding.
2,309 Pulls 8 Tags Updated 7 months ago
This is the continuation of Qwen3 thinking model (MOE), with improved quality and depth of reasoning. (quantized UD-Q4_K_XL, thinking without switching off)
228 Pulls 3 Tags Updated 9 months ago
s1 is a reasoning model finetuned from Qwen2.5-32B-Instruct on just 1,000 examples. It matches o1-preview & exhibits test-time scaling via budget forcing.
558 Pulls 3 Tags Updated 1 year ago
418 Pulls 9 Tags Updated 1 year ago
a reasoning model finetuned from Qwen2.5-32B-Instruct on just 1,000 examples. It matches o1-preview & exhibits test-time scaling via budget forcing. available in [F16, q8_0, q6_K, q4_K_S]
88 Pulls 4 Tags Updated 1 year ago
Custom 3-bit coding model optimized for local agent workflows on 16 GB GPUs. Features a 128K context window for large codebases, long-running tasks, and coding assistants such as OpenCode. Designed for efficient local inference with strong code generation
726 Pulls 1 Tag Updated 3 weeks ago
Validated runtime configuration + measured results for an abliterated (uncensored) Qwen3-Coder-Next 80B-A3B running as an agentic coding backend on CPU via ik_llama.cpp
167 Pulls 1 Tag Updated 1 week ago