13 yesterday

Ornith-1.0-35B is an efficient 35-billion parameter open-source reasoning model optimized for single-GPU deployment and trained via reinforcement learning to achieve state-of-the-art performance in agentic coding and tool-calling tasks.

vision 35b
81cb11a21984 · 218B
{{ if .System }}<|im_start|>system
{{ .System }}<|im_end|>
{{ end }}{{ if .Prompt }}<|im_start|>user
{{ if .Image }}<|image_pad|>{{ end }}{{ .Prompt }}<|im_end|>
{{ end }}<|im_start|>assistant
{{ .Response }}<|im_end|>