Models 106
Latest releases first. Find a model for what you’re making.
0–7d New 8–30d Recent 31d+ Older · days since release
GLM 5.3 FlashX
2dGLM-5.3-FlashX is the high-speed variant of Z.ai's GLM-5.3-Flash, a native multimodal model delivering inference speeds of up to 200 tokens/s.
Open weightsTernary Bonsai 2 27B
2dBonsai 2 27B is a 27B-parameter reasoning model from PrismML derived from Qwen3.8-27B.
405.6K downloads / mo978 likesDeepSeek-V4.1-Flash-NVFP4
4d550 downloads / mo62 likesDeepSeek Flash Latest
6dThis model always redirects to the latest model in the DeepSeek Flash family.
Open weightsDeepSeek-V4.1-Flash-Abliterated-Cybersecurity-Unleashed
7d0 downloads / mo114 likesSwift-Qwen3.8-27B-GGUF
9d100.2K downloads / mo284 likesAgnes-3.0-Flash
9d1.4K downloads / mo228 likesQwen3.8-27B-TWIN-TURBO-Fable-Cold-Fusion-709-L-Uncensored-NM-DAU-NEO-MTP-GGUF
10d163.6K downloads / mo168 likesLing 3.0 Flash VL
10dLing 3.0 Flash VL builds on Ling 3.0 Flash (124B total / 5.5B active MoE from InclusionAI), further strengthening its language capabilities while adding…
7.7K downloads / mo97 likesDeepSeek V4.1 Flash
10dDeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
429.9K downloads / mo3.2K likesDeepSeek-V4.1-Flash-UNCENSORED-FP8
10d33.1K downloads / mo283 likesSignal-3.8-27B-GGUF
11d17.1K downloads / mo89 likesSwift-Qwen3.8-27b
12d6.3K downloads / mo448 likesNex-N2.5-Pro (free)
12dNex-N2.5 is an agentic model built to turn goals into working, verified outcomes.
31.7K downloads / mo665 likesNex-N2.5-Mini (free)
12dNex-N2.5 is an agentic model built to turn goals into working, verified outcomes.
8K downloads / mo836 likesQwen3.8-Flash-Next-GSQ-RCO-GGUF
13d21.2K downloads / mo144 likesHuihui-Qwen3.8-Flash-Next-abliterated-GGUF
14d59.7K downloads / mo76 likesZDTaichu5.0-9B
16d1.8K downloads / mo182 likesQwopus3.8-27B-Flash-GGUF
16d276.4K downloads / mo230 likesQwen3.8 Max (0902)
17dQwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team.
Open weightsQwen3.8-Flash-Next-NVFP4
18d169.2K downloads / mo264 likesGLM-5.3-Flash-NVFP4
18d17.9K downloads / mo101 likesQwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF
19d1.2M downloads / mo913 likesjina-ocr-v1
19d595 downloads / mo93 likesCyber-Tiel-Coder-35B-A3B-GGUF-MTP
20d32.9K downloads / mo96 likesQwen3.8-27B-GSQ-RCO-GGUF
23d1.1M downloads / mo1.4K likesGLM Flash Latest
24dThis model always redirects to the latest model in the GLM Flash family.
Open weightsQwen-Drive-1.0-4B
24d5.4K downloads / mo218 likesQwen3.8 Flash
25dQwen3.8 Flash is a multimodal reasoning model from Alibaba.
724.1K downloads / mo5.4K likesQwen3.8-Flash-Next-Uncensored-GGUF
25d203.4K downloads / mo359 likesQwen3.8-Flash-Next-GGUF
25d1.5M downloads / mo995 likesGLM 5.3 Flash
25dGLM-5.3-Flash is a native multimodal model from Z.ai.
2.7M downloads / mo2.5K likesTiel-Coder-35B-A3B-GGUF-MTP
27d782.8K downloads / mo173 likesDeepSeek V4 Flash Vision Exp
30dDeepSeek V4 Flash Vision Exp is an experimental vision-enabled version of DeepSeek V4 Flash 0731 from DeepSeek, adding image understanding while matching the…
705K downloads / mo922 likesQwen3.8-27B-Uncensored-Cyber-agentic-imatrix-GGUF
32d24.5K downloads / mo146 likesTiel-Coder-35B-A3B-GGUF
32d435.8K downloads / mo306 likesQwen3.8-27B-Uncensored
33d72.5K downloads / mo371 likesQwen3.8-27B-GGUF
33d20.7K downloads / mo99 likesQwen3.8-27B-Uncensored-MLX
34d180.3K downloads / mo1.4K likesQwen3.8-27B-Uncensored-HauhauCS-Aggressive-MTP-GGUF
34d2.3M downloads / mo1.3K likesQwen3.8-27B-Uncensored-GGUF
35d310.3K downloads / mo1.1K likesHuihui-Qwen3.8-27B-abliterated-GGUF
35d2.8M downloads / mo790 likesQwen3.8-27B-Uncensored-FP8
36d428K downloads / mo1.7K likesQwen3.8 27B
37dQwen3.8 27B is an open-weight dense vision-language model from Qwen.
7.4M downloads / mo15.7K likesoccamy-1.0
38d1.3K downloads / mo132 likesMuse Glimmer 30B
42dMuse Glimmer 30B is a dense, open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark and optimized for autonomous agents on…
303.2K downloads / mo1.9K likesInkling Small
52dInkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total.
489.5K downloads / mo408 likesQwen3.7 Flash
55dQwen3.7 Flash is a vision-language reasoning model from Alibaba.
Open weights
