Local AI model library
Browse open and open-weight models by task fit, memory requirement, quantization, runtime, and source confidence.
DeepSeek-R1-Distill 70B
human-review-recommendedmitDeepSeek-R1 distilled to Llama 70B base. Strong general reasoning.
DeepSeek-R1-Distill 32B
human-review-recommendedmitDeepSeek-R1 reasoning model distilled to 32B. Excellent at math & logic.
Qwen3 235B-A22B
human-review-recommendedapache-2.0Alibaba's flagship Qwen3. Competitive with GPT-4 class models.
Phi-4 14B
human-review-recommendedmitMicrosoft's compact 14B. Punches way above its weight, especially on math.
GLM 5.1 FP8
Needs reviewmitZ.ai's GLM 5.1 FP8. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
GLM 4.7 Flash
Needs reviewmitZ.ai's GLM 4.7 Flash. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Kimi K2.7 Code
Needs reviewotherMoonshot AI's Kimi K2.7 Code. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Kimi K2.6
Needs reviewotherMoonshot AI's Kimi K2.6. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Kimi K2 Thinking
Needs reviewotherMoonshot AI's Kimi K2 Thinking. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
MiniMax M3
Needs reviewotherMiniMax's MiniMax M3. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
MiniMax M3 MXFP8
Needs reviewotherMiniMax's MiniMax M3 MXFP8. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
MiniMax M2.7
Needs reviewotherMiniMax's MiniMax M2.7. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Gemma 4 31B It QAT W4a16 Ct
Needs reviewapache-2.0Google's Gemma 4 31B It QAT W4a16 Ct. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Gemma 4 12B It QAT W4a16 Ct
Needs reviewapache-2.0Google's Gemma 4 12B It QAT W4a16 Ct. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Gemma 4 E4B It QAT Q4 0 GGUF
Needs reviewapache-2.0Google's Gemma 4 E4B It QAT Q4 0 GGUF. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
FLUX.2 Dev
Needs reviewotherBlack Forest Labs's FLUX.2 Dev. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
FLUX.2 Klein 9B
Needs reviewotherBlack Forest Labs's FLUX.2 Klein 9B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
FLUX.2 Klein 4B
Needs reviewapache-2.0Black Forest Labs's FLUX.2 Klein 4B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Wan2.2 TI2V 5B
Needs reviewapache-2.0Alibaba's Wan2.2 TI2V 5B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Wan2.2 I2V A14B
Needs reviewapache-2.0Alibaba's Wan2.2 I2V A14B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
LTX 2.3
Needs reviewotherLightricks's LTX 2.3. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
LTX 2.3 FP8
Needs reviewotherLightricks's LTX 2.3 FP8. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Wan2.2 T2V A14B
Needs reviewapache-2.0Alibaba's Wan2.2 T2V A14B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Qwen3.6 27B
Human reviewedapache-2.0Alibaba's open-weight Qwen3.6 27B. Strong coding-agent, reasoning, long-context, and vision-language model with 262K native context.
Qwen3-Coder 30B-A3B
human-review-recommendedapache-2.0Alibaba's Qwen3-Coder 30B with MoE architecture. Top coding model in <30B class.
Qwen2.5-Coder 32B
human-review-recommendedapache-2.0Qwen 2.5 Coder 32B. Predecessor to Qwen3-Coder, still very capable.
GLM 5.2 FP8
Needs reviewmitZ.ai's GLM 5.2 FP8. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Kimi K2.5
Needs reviewotherMoonshot AI's Kimi K2.5. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Kimi K2 Instruct 0905
Needs reviewotherMoonshot AI's Kimi K2 Instruct 0905. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
MiniMax M2.5
Needs reviewotherMiniMax's MiniMax M2.5. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
DeepSeek-Coder-V2 Lite 16B
human-review-recommendeddeepseekDeepSeek's efficient MoE coder. 16B total / 2.4B active.
Llama 3.3 70B
human-review-recommendedllamaMeta's Llama 3.3 70B. Strong all-rounder, 128K context.
MiniMax M2
Needs reviewotherMiniMax's MiniMax M2. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Diffusiongemma 26B A4B It
Needs reviewapache-2.0Google's Diffusiongemma 26B A4B It. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Devstral 24B
human-review-recommendedapache-2.0Mistral's coding-focused model. Strong agent capabilities.
Kimi VL A3B Thinking 2506
Needs reviewmitMoonshot AI's Kimi VL A3B Thinking 2506. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
FLUX.1 Kontext Dev
Needs reviewotherBlack Forest Labs's FLUX.1 Kontext Dev. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Stable Diffusion 3.5 Medium
Needs reviewotherStability AI's Stable Diffusion 3.5 Medium. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Qwen3 30B-A3B
human-review-recommendedapache-2.0Qwen3 30B with MoE. Same family as Coder, general-purpose.
Stable Diffusion 3.5 Large
Needs reviewotherStability AI's Stable Diffusion 3.5 Large. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Wan2.2 S2V 14B
Needs reviewapache-2.0Alibaba's Wan2.2 S2V 14B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Wan2.2 Animate 14B
Needs reviewapache-2.0Alibaba's Wan2.2 Animate 14B. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
LTX 2.3 Nvfp4
Needs reviewotherLightricks's LTX 2.3 Nvfp4. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Mistral Small 3.1 24B
human-review-recommendedapache-2.0Mistral Small 3.1 with vision. Strong multilingual support.
Gemma 4 E2B It QAT Mobile Transformers
Needs reviewapache-2.0Google's Gemma 4 E2B It QAT Mobile Transformers. Auto-imported from Hugging Face seed; review memory and benchmark details before using for final ranking.
Gemma 3 27B
human-review-recommendedgemmaGoogle's open multimodal 27B. Vision + language in one model.
Qwen3 8B
human-review-recommendedapache-2.0Qwen3 8B. Compact general-purpose, runs on any laptop.
Qwen2.5-VL 32B
human-review-recommendedapache-2.0Qwen 2.5 VL (Vision-Language). Top open vision model.
LLaVA-1.6 34B
human-review-recommendedapache-2.0LLaVA 1.6 multimodal model. Image understanding + chat.
Wan2.1 14B
human-review-recommendedapache-2.0Alibaba's Wan 2.1 text-to-video. Open weights, runs on consumer GPUs.
CogVideoX 5B
human-review-recommendedapache-2.0Zhipu AI's CogVideoX 5B. Compact open video model.
BGE-M3 Embedding
human-review-recommendedmitBAAI's BGE-M3. Multilingual embedding for RAG.
Nomic Embed Text v1.5
human-review-recommendedapache-2.0Nomic AI's text embedding. Long context, fully open.
Whisper Large v3
human-review-recommendedmitOpenAI's Whisper Large v3. Top open ASR model.
F5-TTS
human-review-recommendedmitF5-TTS. Open-source TTS with voice cloning.
CosyVoice 300M
human-review-recommendedapache-2.0Alibaba's CosyVoice. Multilingual TTS with emotion control.
SmolLM2 1.7B
human-review-recommendedapache-2.0Hugging Face's SmolLM2 1.7B. Tiny but capable, fits anywhere.
FLUX.1-dev
human-review-recommendedapache-2.0Black Forest Labs' state-of-the-art text-to-image model. Top of GenEval.
Stable Diffusion XL
human-review-recommendedopenrailStability AI's classic SDXL. Best bang/buck for image gen.
HunyuanVideo 13B
human-review-recommendedtencentTencent's open HunyuanVideo. 13B params, high quality.
Llama 3.2 3B
human-review-recommendedllamaMeta's tiny Llama 3.2 3B. Runs on phones & weak hardware.
Gemma 3n 4B
human-review-recommendedgemmaGoogle's Gemma 3n. Multimodal (text/vision/audio) at 4B size.