xM_My
@NullBurn

Local LLM

 DGX Spark 2x
Joined June 2013
Replying to @J37270031
3
1,645
prefill 2300tps @ 32k
4
1
11
2,702
Replying to @ivanfioravanti
- 117.2 tok/s on prose, single stream - 215.8 tok/s on prose, 4 concurrent 2 dgx sparks with NVIDIA Flare & NVPF4 #NVIDIA
2
1
251
Running GLM-5.3-Flash on 2× DGX Spark with NVIDIA Flare ✨ - 1M context window, 2.32M-token shared KV pool - Prose decode, single stream: 117.2 tok/s - Prose decode, 2 concurrent: 167.4 tok/s aggregate - Prose decode, 4 concurrent: 215.8 tok/s aggregate - 32K prefill: 2,327.9 tok/s - Available memory after loading: ~13 GiB on head /worker each Checkpoint: GLM-5.3-Flash-NVFP4-MXFP8-CSF-QAD github.com/NVIDIA/flare-loca…
Mia
@MiaAI_lab
Oct 21
NVIDIA just released Flare, a local inference server built for DGX Spark and RTX Blackwell GPUs - Built for long chats, agents and tool calls - Serves from Rust, no Python - Plans memory up front - FP8 KV cache by default Interesting timing... right around when DeepSeek open-sourced their inference engine. Link below.
66
27
270
43,078
Replying to @uzairansar
I have tried other mobile apps for Hermes. They use the Hermes origin desktop gateway, but the problem is that you can't use the same session which is opening on the desktop app. Did you solve the problem?
1
1
79
Replying to @ashxhart @aijoey
where can I get the QR code？
1
70
Replying to @llqoli
为什么000222444无法订阅？
12
This is how and when I will die tinyurl.com/howwillyoudie2 .
1
#Splatoon2 #NintendoSwitch
@twostraws hi,Paul I got a error when I follow your project1 tutorial---Listing images with NSFileManager.
1
Load more