FRAMEWIREIndonesiaUpdated Aug 17Live wire
0:00 / 0:00

Gemini 3.7 Flash vs Qwen 3.8 vs Grok 4.6 vs GLM 5.3

0xMarioNawfalAug 1733
0:00 / 0:00

Qwen 3.8 27B vs. DeepSeek V4 Flash 0731

Even though DeepSeek is 10x the size, Qwen actually produced better results

bartslodyczkaAug 16
0:00 / 0:00

Awesome! Qwen-3.8-27B NVFP4 quantization, enable MTP

The throughput speed on DGX Spark has reached 17tok/s, nearly doubled again If coupled with SGLang’s MTP, wouldn’t it be possible to take off✈️😄

LonelyAug 163
0:00 / 0:00

You are paying Opus 4.8 prices for steps a 27B model finishes just as well.

NVIDIA shipped an open source router this week, NeMo SwitchYard. It reads the task and picks the model. Their own numbers: SwitchYard across Opus 4.8 plus smaller models completed more tasks than Opus

godgivenAug 166