Spark-Bench

All projects

Apps & Tools

A parallel LLM hardware stress-test and benchmarking dashboard

byWësche

Open source

Spark-Bench media is blocked

Allow external media to connect to the provider and play this content.

Description

A technical benchmarking tool designed to visualize and measure the throughput of local LLM hardware under heavy load. It demonstrates high-concurrency performance, such as running Qwen3.8 models across 64 parallel users on mining-focused GPUs like the NVIDIA CMP 170HX using speculative decoding and vLLM.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.

Details

Why this is interesting: Showcases impressive local inference throughput and context capacity by repurposing mining hardware for…
Attributed to: Wësche@weschenex1q
Built with: Ling 3.0 Flash Fin Qwen3.8-Flash-Next Qwen3.8-27B Gemini 3.7 Flash Qwen3.7 Flash 3.5 Flash Cyber 3.6 Flash Step 3.7 Flash Gemini 3.5 Flash 2.5 Flash Qwen3 8B Qwen3 14B Qwen3 30B-A3B
Links: View tweet
Tags: #benchmark #hardware #inference #visualization

Discovered: X
First posted: X · Sep 1, 2026
Listed: Added 15d ago
Build: Moderate
Models: Strong

Spark-Bench

Description

Timeline

Spark-Bench

Description

Timeline