Self-hosted DeepSeek inference beats every vendor on price/perf, boosting the open-weight enterprise case
bendee983 · x · 2026-10-19
thdxr's team reports their first test run of a new DeepSeek inference setup on a fresh GPU rack achieves better price/performance than every vendor they've found, with more optimization to come. bendee983 argues this is the open-weight advantage: self-host, optimize inference for specific workloads, and fine-tune on proprietary UX data — predicting enterprises will run open-weight models for 80-90% of tasks and reserve closed frontier models for edge cases.
More from Infra
- LLM Quantization Explained: Bit Width Alone Doesn't Determine Accuracy or Speed — techNmak · 2026-10-19
- 303 companies mapped on Nebius: from 1X robots to Mastercard, tracked in independent atlas — pdamodaran · 2026-10-19
- vLLM details MiniMax M3.1 Flash serving stack: 1.7x faster indexer decode via NVFP4 — vllm_project · 2026-10-19
- DeepSeek V4.1 Flash at 2.61 bpw hits 26 tok/s on dual RTX 3090s with dsx — whiteh4cker · 2026-10-19
- Neocloud shakeout ahead: only 7-10 of hundreds of AI clouds may survive, Nvidia most exposed — The Decoder · 2026-10-19
- Wan 2.2 I2V on a 12GB GPU uses only 8-9GB VRAM—system RAM is the real bottleneck — noise_to_motion · 2026-10-19