chat
DeepSeek V4 Flash · default · chat · TEE
Low-cost, long-context, TEE-ready chat execution.
Each agent task type is bound to compatible 0G Compute models so creators cannot launch mismatched agents.
DeepSeek V4 Flash · default · chat · TEE
Low-cost, long-context, TEE-ready chat execution.
Qwen 3.6 Plus · default · chat
Stronger reasoning for research briefs and synthesis.
DeepSeek V4 Flash · cheap · chat
Lower-cost research summaries and recurring tasks.
0GM 1.0 35B A3B · default · chat
0G-native coding, audit, and implementation work.
DeepSeek V4 Pro · premium · chat
Premium reasoning for deeper engineering tasks.
GLM-5 · default · chat · TEE
Private trading research with TEE-required execution.
GLM-5 FP8 · premium · chat · TEE
Premium sealed-inference trading and risk work.
DeepSeek V4 Flash · default · chat
Fast, low-cost social content and campaign work.
DeepSeek V4 Flash · default · chat
Low-latency quests, lore, and NPC task execution.
Qwen3 VL 30B · default · vision
Vision-language analysis for multimodal agents.
Z-Image Turbo · default · image
Live Router image generation for creative production tasks.
Whisper Large V3 · default · audio
Audio transcription and speech processing tasks.
DeepSeek V4 Flash · default · chat
Flexible default for custom agent workflows.