AI Model Hub

Browse all available AI models with support for multiple inference frameworks

Model Replacement Deployment Guide
Total Models 105
Filtered 105
Available 105
Showing 97-105 / 105 models
Page 5 / 5
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_128k_b1_2core_1chip
27 gguf w4a8 M50 131,072
Text Text

24G卡可以跑

Detail Download
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_128k_b1_2core_1chip(dynamic)
27 gguf w4a8 M50 131,072 Vision
Text Text

Type: Qwen3.8 | Size: 27 | Quant: w4a8 | Context: 131072 | Chip: M50 | Build: 1.5.0 | Status: Temporary

Detail Download
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_256k_b1_2core_1chip
27 gguf w4a8 M50 262,144
Text Text
HLIEvLLM HLIELLama

需要24G+的卡

Detail Download
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_256k_b1_2core_2chips
27 gguf w4a8 M50 262,144
Text Text
HLIELLama HLIEvLLM

Type: Qwen3.8 | Size: 27 | Quant: w4a8 | Context: 262144 | Chip: M50 | Build: 1.5.0 | Status: Stable

Detail Download
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_256k_b1_2core_2chips(dynamic)
27 gguf w4a8 M50 262,144 Vision
Text Text

Type: Qwen3.8 | Size: 27 | Quant: w4a8 | Context: 262144 | Chip: M50 | Build: 1.5.0 | Status: Temporary

Detail Download
QWEN3.8
QWEN3.8 27
Qwen3.8_27b-mtp_256k_b1_2core_4chips
27 gguf w4a8 M50 262,144
Text Text
HLIEvLLM HLIELLama

Type: Qwen3.8 | Size: 27 | Quant: w4a8 | Context: 262144 | Chip: M50 | Build: 1.5.0 | Status: Stable

Detail Download
OpenAI
WHISPER 0.8
Whisper_0.8b__b1_2core_1chip
0.8 gguf w4a8 M50
Audio Text
HLIECpp HLIEPython HLIEvLLM

Type: Whisper | Size: 0.8 | Quant: w4a8 | Chip: M50 | Build: 1.4.0 | Status: Stable

Detail Download
WHISPER-TURBO
WHISPER-TURBO 0.8
Whisper-Turbo_0.8b__b1_2core_1chip
0.8 gguf w4a8 M50
Audio Text
HLIECpp HLIEPython HLIEvLLM

Type: Whisper-Turbo | Size: 0.8 | Quant: w4a8 | Chip: M50 | Build: 1.4.0 | Status: Stable

Detail Download
Z-IMAGE-TURBO
Z-IMAGE-TURBO 6
Z-Image-Turbo_6b_4k_b1_2core_1chip
6 gguf w8a8 M50 4,096
Text Text

Type: Z-Image-Turbo | Size: 6 | Quant: w8a8 | Context: 4096 | Chip: M50 | Build: 1.4.0 | Status: Testing

Detail Download