Osaurus-AI

Osaurus-AI's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a model 1 day ago

OsaurusAI/Raptor-0.6.1-preview-GGUF

Text Generation • 4B • Updated 1 day ago • 266

published a model 1 day ago

OsaurusAI/Raptor-0.6.1-preview-GGUF

Text Generation • 4B • Updated 1 day ago • 266

updated a model 2 days ago

OsaurusAI/Raptor-0.6.1-preview-JANG_6M

Text Generation • 4B • Updated 2 days ago • 200

published a model 2 days ago

OsaurusAI/Raptor-0.6.1-preview-JANG_6M

Text Generation • 4B • Updated 2 days ago • 200

published a model 3 days ago

OsaurusAI/Raptor-0.6-preview-JANG_6M

Text Generation • 4B • Updated 3 days ago • 66

updated 3 models 4 days ago

OsaurusAI/Raptor-0.6-preview-JANG_6M

Text Generation • 4B • Updated 3 days ago • 66

OsaurusAI/Spark-X2.5-4B-JANG_6M

Text Generation • 4B • Updated 4 days ago • 505

OsaurusAI/MiniCPM5-2B-JANG_8M

Text Generation • 3B • Updated 4 days ago • 940 • 1

published a model 4 days ago

OsaurusAI/Spark-X2.5-4B-JANG_6M

Text Generation • 4B • Updated 4 days ago • 505

published a model 5 days ago

OsaurusAI/MiniCPM5-2B-JANG_8M

Text Generation • 3B • Updated 4 days ago • 940 • 1

updated a model 7 days ago

OsaurusAI/Raptor-v0.5-8B-A1B-JANG_6M

Text Generation • 6B • Updated 7 days ago • 2.26k • 11

updated 5 models 9 days ago

OsaurusAI/Qwen3.8-27B-JANG_6D

Image-Text-to-Text • 28B • Updated 9 days ago • 979 • 2

OsaurusAI/Qwen3.8-27B-JANG_4D

Image-Text-to-Text • 28B • Updated 9 days ago • 918

OsaurusAI/Qwen3.8-27B-JANG_2D

Image-Text-to-Text • 28B • Updated 9 days ago • 824

OsaurusAI/Qwen3.8-Flash-Next-JANG_2L

Image-Text-to-Text • 180B • Updated 9 days ago • 809

OsaurusAI/Qwen3.8-Flash-Next-JANG_1L

Image-Text-to-Text • 177B • Updated 9 days ago • 1.4k • 4

New activity in JANGQ-AI/GLM-5.3-FP8 11 days ago

fix: default clear_thinking=true in chat template + add repetition_penalty=1.1 default — mitigates known GLM-5.3 reasoning-loop degeneration on multi-turn agent workloads (SGLang #36669, opencode #45533). Upstream Z.ai default `clear_thinking=false` replays prior reasoning into subsequent turns, priming a decode attractor. Z.ai's own docs recommend `clear_thinking=true` for chat scenarios — this bakes it in.

#1 opened 11 days ago by

Osaurus-AI

New activity in JANGQ-AI/GLM-5.3-Flash-W4A16 11 days ago

fix: default clear_thinking=true in chat template + add repetition_penalty=1.1 default — mitigates known GLM-5.3 reasoning-loop degeneration on multi-turn agent workloads (SGLang #36669, opencode #45533). Upstream Z.ai default `clear_thinking=false` replays prior reasoning into subsequent turns, priming a decode attractor. Z.ai's own docs recommend `clear_thinking=true` for chat scenarios — this bakes it in.

#1 opened 11 days ago by

Osaurus-AI

New activity in JANGQ-AI/GLM-5.3-W4A16 11 days ago

fix: default clear_thinking=true in chat template + add repetition_penalty=1.1 default — mitigates known GLM-5.3 reasoning-loop degeneration on multi-turn agent workloads (SGLang #36669, opencode #45533). Upstream Z.ai default `clear_thinking=false` replays prior reasoning into subsequent turns, priming a decode attractor. Z.ai's own docs recommend `clear_thinking=true` for chat scenarios — this bakes it in.

#1 opened 11 days ago by

Osaurus-AI

New activity in JANGQ-AI/GLM-5.3-Flash-JANG-MTP 11 days ago

fix: default clear_thinking=true in chat template + add repetition_penalty=1.1 default — mitigates known GLM-5.3 reasoning-loop degeneration on multi-turn agent workloads (SGLang #36669, opencode #45533). Upstream Z.ai default `clear_thinking=false` replays prior reasoning into subsequent turns, priming a decode attractor. Z.ai's own docs recommend `clear_thinking=true` for chat scenarios — this bakes it in.

#1 opened 11 days ago by

Osaurus-AI