Abliterated (Uncensored) GLM4.6 Flash
68 Pulls 11 Tags Updated yesterday
8 Pulls 1 Tag Updated 3 days ago
2 Pulls 1 Tag Updated 3 days ago
7 Pulls 1 Tag Updated 4 days ago
28 Pulls 1 Tag Updated 4 days ago
8 Pulls 1 Tag Updated 4 days ago
1 Tag Updated 4 days ago
10 Pulls 1 Tag Updated 4 days ago
5 Pulls 3 Tags Updated 5 days ago
Abliterated (Uncensored) Devstral Small 2
225 Pulls 12 Tags Updated 5 days ago
A state-of-the-art mixture-of-experts (MoE) language model. Kimi K2-Instruct-0905 demonstrates significant improvements in performance on public benchmarks and real-world coding agent tasks.
119 Pulls 6 Tags Updated yesterday
The first installment in the Qwen3-Next series with strong performance in terms of both parameter efficiency and inference speed.
1,221 Pulls 10 Tags Updated 1 week ago
Abliterated Qwen3-14B with 80% reduced refusals while preserving coherence (KL 0.98)
389 Pulls 6 Tags Updated 5 days ago
Advanced agentic, reasoning and coding capabilities.
148 Pulls 4 Tags Updated 1 week ago
Orchestrator-8B is a state-of-the-art 8B parameter orchestration model designed to solve complex, multi-turn agentic tasks by coordinating a diverse set of expert models and tools.
398 Pulls 5 Tags Updated 2 weeks ago
Fara-7B is Microsoft's first agentic small language model (SLM) designed specifically for computer use.
1,024 Pulls 5 Tags Updated 3 weeks ago
DANGER: CONVERSATION MAY DEVIATE FROM SAFE DESPITE HARMLESS PROMPTS. Embedded use only. NOT intended as-is for end-users.
97 Pulls 1 Tag Updated 3 weeks ago
DANGER: CONVERSATION MAY DEVIATE FROM SAFE DESPITE HARMLESS PROMPTS. Embedded use only. NOT intended as-is for end-users
73 Pulls 1 Tag Updated 3 weeks ago
MiroThinker v1.0 is an open-source research agent designed to advance tool-augmented reasoning and information-seeking capabilities.
739 Pulls 18 Tags Updated 3 weeks ago
Flagship vision-language model of Qwen and also a significant leap from the previous Qwen2-VL.
885 Pulls 16 Tags Updated 1 month ago