ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-Coder-GGUF Image-Text-to-Text ⢠117B ⢠Updated 13 days ago ⢠861k ⢠551
view post Post 4,394 š Introducing Halo 1.0Today, we are open-sourcing Halo, the training framework we use to train every model at White Circle.It comes with: š§ Full post-training stack: SFT, DPO/KTO/SMPO, reward modeling, GRPO, distillationš¤ Async multi-turn RL with vLLM/SGLang rollouts and sandboxed tool useā” ~2.8Ć TRL throughput on 8Ć B300 (EP+FSDPv2, FA4, fp8/fp4)š¤ Dense HF models + 15 MoE families (Qwen, GLM, Mistral, DeepSeek-V4ā¦)š ļø One halo command, prebuilt Docker images, and docs for humans and agentsš» https://github.com/whitecircle/haloTry it and tell us what you're training See translation 1 reply Ā· š 9 9 ā¤ļø 4 4 + Reply
dealignai/Bonsai-2-27B-Ternary-CRACK-GGUF Text Generation ⢠27B ⢠Updated Sep 18 ⢠25.8k ⢠169
prism-ml/Ternary-Bonsai-2-27B-gguf Text Generation ⢠27B ⢠Updated 2 days ago ⢠1.85M ⢠2.89k
peculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF Image-Text-to-Text ⢠35B ⢠Updated 7 days ago ⢠308k ⢠418
ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF Image-Text-to-Text ⢠27B ⢠Updated Sep 2 ⢠1.31M ⢠2.49k
peculiar-ragdoll/Cyber-Tiel-Coder-35B-A3B-GGUF Image-Text-to-Text ⢠35B ⢠Updated 6 days ago ⢠72.1k ⢠145
AnonimousA/Qwen3.8-Flash-Next-REAP-320-GGUF Text Generation ⢠132B ⢠Updated Sep 9 ⢠3.56k ⢠27
deepseek-ai/DeepSeek-V4.1-Flash Image-Text-to-Text ⢠763B ⢠Updated 17 days ago ⢠2.13M ⢠⢠4.83k