|
Cannot reach 200Gbps with 4x DGX Spark through MikroTik CRS 812 switches
|
|
10
|
249
|
October 14, 2026
|
|
veloGB10 v0.7.0 RELEASE - Qwen3.8-Flash-Next TP=2 @ 186 tok/sec , SINGLE at 137 tok/sec on code
|
|
49
|
2996
|
October 14, 2026
|
|
Your DGX Spark as a production AI server: one command, a cockpit, and an agent in the browser
|
|
8
|
794
|
October 14, 2026
|
|
Qwen3.8-27B Benchmark Study on DGX Sparks: How speculative decoding results stack up
|
|
0
|
128
|
October 13, 2026
|
|
Prioritizing Compute on the GB10
|
|
10
|
412
|
October 13, 2026
|
|
The adventures of Nemohermes (Nemoclaw)
|
|
1
|
174
|
October 13, 2026
|
|
New DGX Spark Production?
|
|
8
|
1506
|
October 12, 2026
|
|
Pixal3D (image-to-3D) on DGX Spark: what it takes on GB10, and why low_vram should be off
|
|
0
|
135
|
October 12, 2026
|
|
Hermes is losing its glory, what is next?
|
|
92
|
5725
|
October 12, 2026
|
|
Dgxtop v2: unified memory visualization and GPU process history for DGX Spark
|
|
0
|
147
|
October 12, 2026
|
|
Commercially Available Rack/Cooling for 1+N Sparks?
|
|
8
|
1525
|
October 12, 2026
|
|
Three times ( VoiceClone | VoiceDesign | CustomVoice ) - Faster-Qwen3-TTS for NVIDIA DGX Spark (GB10)
|
|
63
|
4110
|
October 11, 2026
|
|
60 tok/s GLM-5.3 Flash on a single DGX Spark
|
|
16
|
5845
|
October 11, 2026
|
|
Ninfer-gb10
|
|
0
|
260
|
October 10, 2026
|
|
Athena's Engine: DeepSeek and Qwen3.8 Next — 1,126 tokens/s prefill, 32 tokens/s decode, 256k ctx on 1DGX Spark
|
|
12
|
1594
|
October 10, 2026
|
|
GLM-5.3-Flash NVFP4 on 3x DGX Spark, stock DGX OS: ~3,300 tok/s prefill, 3M-token KV, up to 64 concurrent requests (kindling TP=3)
|
|
7
|
771
|
October 10, 2026
|
|
HOW-TO: setup-dgx-spark docker inference - A "Sane" Inference Stack for GB10 (Need Contributors!)
|
|
40
|
3737
|
October 10, 2026
|
|
Spark-comfyui, a self-healing ComfyUI setup for DGX Spark
|
|
35
|
5180
|
October 10, 2026
|
|
Introducing Tool Eval Bench CLI
|
|
214
|
10669
|
October 10, 2026
|
|
Qwen-Image-2.1 via vLLM-OMNI
|
|
4
|
996
|
October 9, 2026
|
|
Running EmbeddingGemma2 on DGX Spark
|
|
3
|
274
|
October 8, 2026
|
|
Qwen-Image-2.1 on DGX Spark: 54 s per 1024² image (40 steps), 31.6 GB, and a fix for the pink vertical line
|
|
1
|
497
|
October 8, 2026
|
|
Qwen3.8-Flash-Next on 3 DGX Sparks — 3 Recipes, Full Debugging Journey, Final Reproducible Setup
|
|
0
|
235
|
October 7, 2026
|
|
VibeVoice-ASR on DGX Spark: a 59-minute recording in one pass, with speakers and timestamps (no chunking)
|
|
0
|
169
|
October 7, 2026
|
|
Switchless 2x 4x 6x GB10 Clusters Serving GLM, DeepSeek, and Qwen
|
|
25
|
2338
|
October 6, 2026
|
|
[Community project] Thinkube: Kubernetes done right on your own Spark
|
|
0
|
128
|
October 5, 2026
|
|
Single Spark Qwen3.8 Flash-Next + Qwen3.8 27B Left/Right Brain Hermes MoA (35-39 toks/s)
|
|
23
|
2526
|
October 4, 2026
|
|
Full GLM-5.3 (753B) on 4x DGX Spark - 3-bit EXL3 quant + TensorFold TP4 recipe
|
|
1
|
401
|
October 4, 2026
|
|
6x GB10 Cluster w/ MikroTik CRS812 = 768GB RAM
|
|
8
|
1741
|
October 3, 2026
|
|
Paris Nvidia Agent claw contest demo
|
|
0
|
140
|
October 3, 2026
|