@TheAhmadOsman: Laguna S 2.1 118B-A8B on DGX Station using - NVFP4 - FP8 KV Cache Can run 10 parallel agents > with 256k context each >…
Summary
Laguna S 2.1 118B-A8B model runs on DGX Station with NVFP4 and FP8 KV Cache, achieving ~1k tokens/second for 10 parallel agents with 256k context each.
View Cached Full Text
Cached at: 07/25/26, 06:01 AM
Laguna S 2.1 118B-A8B on DGX Station using
- NVFP4
- FP8 KV Cache
Can run 10 parallel agents
> with 256k context each > at ~1k tokens/second
Some other numbers below https://t.co/Kb7RaTxBQz
Similar Articles
Laguna-S-2.1 runs on my 6 years old gaming PC!
Laguna-S-2.1, a quantized AI model, runs on a 6-year-old gaming PC with an RTX 3080, achieving 10 t/s decode and using 8.3 GB VRAM and 52.2 GB host RAM.
GLM-5.3-Flash @ DGX Station GB300: ~206 tok/s (single stream), 1M context
The user benchmarks the GLM-5.3-Flash AI model on a DGX Station, achieving ~206 tokens per second in single-stream inference with a 1 million context window, and shares a Docker command for setup.
@no_stp_on_snek: Ran this on Laguna S 2.1 in Poolside's own agent (pool), pointed at a local instance on a DGX Spark, using the prompt l…
A comparison of two AI coding agents building a Mario game: Laguna S 2.1 in Poolside's agent took 62 minutes with self-correction and passed tests, while a previous Qwen model took hours and needed human help; highlights oracle discipline and native harness advantages.
@0xSero: GLM-5.1-478B-NVFP4 Running on: - 4x RTX Pro 6000 - Sglang - 370,000 max tokens (1.75x full context) - p10 27.7 | p90 45…
A quantized 478B-parameter GLM-5.1 model runs on 4×RTX Pro 6000 GPUs via SGLang, delivering 370k-token context at up to 45 tok/s decode and 1340 tok/s prefill, and is demoed driving Figma.
@onusoz: 16x parallel Gemma-4-26B-A4B-NVFP4 runs 18 output tokens/s, aggregate 300 tok/s 1 DGX Spark with 128 GB unified memo…
@onusoz demonstrates running 16 parallel instances of NVIDIA's quantized Gemma-4-26B-A4B-NVFP4 model on a single DGX Spark with 128GB unified memory, achieving 300 tok/s aggregate, showcasing high concurrency without flashinfer.