|
MiniMax M2.7 TQ3 - A TurboQuant 3-bit quantized version of MiniMax-M2.7 for single DGX Spark
|
|
3
|
2002
|
April 27, 2026
|
|
MiniMax M2.7 NFVP4 Recipe & Benchmarks
|
|
69
|
5050
|
April 27, 2026
|
|
Introducing PrismaQuant
|
|
125
|
2942
|
April 27, 2026
|
|
DGX Spark Performance Degradation - GPU Power Draw Issue
|
|
30
|
1629
|
April 27, 2026
|
|
MiMo-V2.5 (New model)
|
|
3
|
154
|
April 27, 2026
|
|
Qwen3.5-122B-A10B on single Spark: up to 51 tok/s (v2.1 — patches + quick-start + benchmark)
|
|
345
|
11045
|
April 27, 2026
|
|
SparkD: The missing dashboard for spark-vllm-docker
|
|
4
|
143
|
April 27, 2026
|
|
Qwen3.6-27B is out!
|
|
46
|
6688
|
April 27, 2026
|
|
Introducing vLLM-Tune — Kernel tuning CLI for vLLM on DGX Spark
|
|
4
|
135
|
April 27, 2026
|
|
Deepseek V4 released
|
|
69
|
5081
|
April 27, 2026
|
|
Qwen/Qwen3.6-35B-A3B (and FP8) has landed
|
|
164
|
13101
|
April 27, 2026
|
|
Three node Spark clusters (without a switch) are now supported in spark-vllm-docker and sparkrun!
|
|
10
|
811
|
April 27, 2026
|
|
GPU PD Throttle Check Tool
|
|
5
|
396
|
April 27, 2026
|
|
NCCL all_gather Performance Halved on Dual Spark Setup (ConnectX-7) After MSI Firmware Update - Solved via Downgrade
|
|
1
|
71
|
April 27, 2026
|
|
Tools mod error in recipe gemma4-26b-a4b after pulling latest spark-vllm-docker
|
|
6
|
139
|
April 27, 2026
|
|
Qwen3.6-27B-Dflash link
|
|
21
|
1163
|
April 27, 2026
|
|
Cloning issue with the AI Workbench Tutorial
|
|
5
|
265
|
April 27, 2026
|
|
Dual Spark Ducted Cooling Cage
|
|
33
|
1126
|
April 27, 2026
|
|
GB10 Hardware Baseline — First Direct Measurements and Findings
|
|
8
|
390
|
April 27, 2026
|
|
Unsloth Studio - semi-manual install
|
|
1
|
135
|
April 27, 2026
|
|
How to fix CPU frequency in DGX Spark
|
|
4
|
271
|
April 27, 2026
|
|
I purchased two DGX Spark Units in March and one has a faulty power supply
|
|
12
|
362
|
April 27, 2026
|
|
Why Turboquant saves DGX twice
|
|
114
|
9303
|
April 27, 2026
|
|
DGX Spark / sm121: silent SDPA `EFFICIENT_ATTENTION` corruption in a custom PyTorch build — diagnostic chain, standalone reproducer, workaround
|
|
0
|
32
|
April 27, 2026
|
|
HOW-TO: setup-dgx-spark docker inference - A "Sane" Inference Stack for GB10 (Need Contributors!)
|
|
37
|
1893
|
April 27, 2026
|
|
DDTree plus diffusion drafting (DFlash) to optimize GB10
|
|
2
|
525
|
April 27, 2026
|
|
Running a Full LLM Stack on DGX Spark GB10 (Your Application -> LiteLLM -> llama-swap -> vLLM / llama.cpp / Ollama)
|
|
10
|
604
|
April 27, 2026
|
|
I keep failing to install Flash Attention 3 in the LTX-2 UV environment
|
|
9
|
608
|
April 26, 2026
|
|
My DGX Spark extra cooling solution
|
|
13
|
548
|
April 26, 2026
|
|
Bfloat16 Quality = Speed?
|
|
42
|
1506
|
April 26, 2026
|