|
RTX PRO 4000 Blackwell: FP16/FP8/FP6/FP4 throughput with FP32 accumulation?
|
|
1
|
46
|
September 11, 2026
|
|
How should we tune Jetson AGX Thor for a CPU-bound web server workload (700 concurrent users), and can the GPU contribute anything?
|
|
1
|
35
|
September 11, 2026
|
|
Switchless 2x 4x 6x GB10 Clusters Serving GLM, DeepSeek, and Qwen
|
|
20
|
1401
|
September 7, 2026
|
|
Motif-3 315B core on one DGX Spark: 83.56 GiB, 316.7 pp / 16.5 tg, reproducible build
|
|
1
|
224
|
September 5, 2026
|
|
Qwen3.8-27B on single DGX Spark: 1.9x stock NVFP4 (2.5x FP8) at 16 concurrent, ~50 tok/s on a single request
|
|
7
|
1087
|
September 1, 2026
|
|
DGX Spark GB10: hard freeze under sustained load (RCU stall on CPU 11), watchdog + kdump both fail — working pstore-only crash capture recipe + eviden
|
|
0
|
81
|
August 28, 2026
|
|
Qwen 3.8 27B + DFlash2
|
|
12
|
4126
|
August 27, 2026
|
|
Cold plate thermal benefit of sub-45°C coolant (down to 4°C) — junction temp / throttling / lifespan data?
|
|
0
|
42
|
August 25, 2026
|
|
Qwen3.8-27B-NVFP4 @ TP=4 running at >80 tok/s sustained & peak at 92 tok/s
|
|
8
|
1427
|
August 23, 2026
|
|
Green-context SM provisioning vs the primary context
|
|
2
|
159
|
August 14, 2026
|
|
Jetpack7.2 T5000 性能测试问题
|
|
7
|
212
|
August 4, 2026
|
|
Is there a particular order to how warps are statically assigned to SM sub partitions?
|
|
2
|
271
|
July 28, 2026
|
|
FUSA Coe on AGX Thor - unexpected results
|
|
1
|
86
|
July 27, 2026
|
|
Accelerated vs non-accelerated HSB latency
|
|
3
|
138
|
July 24, 2026
|
|
Nvfortran OpenACC performance regression: WENO5 reconstruction kernel ~1.7x slower since 24.7 (24.5 good, 24.7 first bad)
|
|
2
|
103
|
July 16, 2026
|
|
AVRCK 3.0: Reducing Inference Cost by Routing Causal Gaps Across Local and Cloud Models
|
|
0
|
87
|
July 14, 2026
|
|
Slow wi-fi on orin nano devkit (RTL8822CE 802.11ac)
|
|
11
|
325
|
June 30, 2026
|
|
GB10 really does hit ~1 PFLOP NVFP4 (2:4 sparse) — measured, with an open-source tool to reproduce it
|
|
22
|
2395
|
June 25, 2026
|
|
Flux.2 Klein 9B on DGX Spark: 2.5x Faster Inference and 59% Lower VRAM with Vitoom Nunchaku
|
|
0
|
443
|
June 25, 2026
|
|
Qwen3.5-122B-A10B on single Spark: up to 51 tok/s (v2.1 — patches + quick-start + benchmark)
|
|
434
|
28534
|
June 24, 2026
|
|
DGX Spark Performance Degradation - GPU Power Draw Issue
|
|
69
|
5431
|
June 15, 2026
|
|
Just another ASUS GX10 NCCL all_gather_perf thread... mpirun... please read if you have an ASUS model multinode setup
|
|
4
|
798
|
May 23, 2026
|
|
Lossless 7.67× LoRA / 8.35× Full FT speedup for Qwen3.5 on DGX Spark (GB10, sm_121a)
|
|
3
|
629
|
May 20, 2026
|
|
T5000 - Sample to use full performance
|
|
3
|
253
|
May 11, 2026
|
|
Dual DGX Spark: NCCL capped at 2.80 GB/s + ib_write_bw crashes at 128KB syndrom 0x88 — matches thread 366266 with additional RoCE degradation
|
|
1
|
331
|
April 20, 2026
|
|
GPU needs to be "warmed up" to achieve maximum performance
|
|
21
|
998
|
April 20, 2026
|
|
NeuralForge GPU Native Knowledge Intelligence Platform Built on DGX Spark GB10
|
|
4
|
404
|
April 17, 2026
|
|
NCCL bandwidth capped at 3 GB/s, GPU PCIe topology reports Gen1 x1 on DGX Spark FE
|
|
5
|
631
|
April 14, 2026
|
|
Qwen3.5 Flash Attention performance inconsistencies
|
|
1
|
566
|
April 5, 2026
|
|
Latest Update (20Mar 2026) on Nvidia Spark FE caps GPU performance
|
|
8
|
891
|
March 20, 2026
|