DeepSeek v4 Flash Hits 82 tok/s on Dual DGX Sparks
andrewchen · x · 2026-08-02
A developer shared benchmarks running DeepSeek v4 Flash (0731) on two NVIDIA DGX Sparks. Using official FP8 weights, single-stream throughput reaches 82 tok/s (peaking at 95 tok/s), and 135 tok/s across 3 concurrent sessions. The deployment recipe is open-sourced.
Related event: DeepSeek V4 Flash Sparks a Wave of Local Deployment Tests(26 posts)→
More from Infra
- Same 1,000 Cloudflare Durable Objects: $4,050/month cut to $22 with one trick — threepointone · 2026-10-16
- NVIDIA ships FP4-quantized Qwen3.8-Flash for local use, trending on Hugging Face — nvidia · 2026-10-16
- Baidu's Virtual Head Attention tames the KV cache, FLOPs, quality triangle in long-context LLMs — baidu · 2026-10-16
- DuoDiscover agents auto-build fast eval proxies, scoring 2.68x TPU kernel speedup — berkeley_ai · 2026-10-16
- Google Raises 2026 Capital Spending to Up to $205 Billion — emmanuelvivier · 2026-10-16
- Google open-sources AX, a Kubernetes-style orchestrator for billions of AI agent tasks — thisdudelikesAI · 2026-10-16