Engine status

TLearth One Chat

tlearth-one-chat · Qwen2.5-32B-Instruct (AWQ)

● Fallback — managed multi-model routing

TLearth Two Code

tlearth-two-code · Qwen2.5-Coder-32B-Instruct (AWQ)

● Fallback — managed multi-model routing

TLearth One Vision

tlearth-one-vision · Qwen2.5-VL-7B-Instruct

● Fallback — managed multi-model routing

TLearth One Embed

tlearth-one-embed · BGE-M3

● Fallback — managed multi-model routing

TLearth One Voice

tlearth-one-voice · Whisper large-v3-turbo + Kokoro-82M

● Not deployed

Sovereign GPUs scale to zero, so the first request after idle can take 30–90 seconds while the model loads.