DeepSeek Uses More Tokens but is Faster per Single Request
Sentdex · x · 2026-07-03
Sentdex observed that DeepSeek consumes about 2.5x more tokens than GLM 5.2 IQ4 but is 5x faster at batch size=1. Once concurrency is introduced, vLLM's concurrent processing capabilities take off rapidly. He also noticed a performance drop in vLLM tensor parallelism at batch size=2 before it recovers, and asked the community for insights on this behavior.
More from Infra
- Moonshot’s Kimi K3 lands on Together with reserved throughput and 65% lower cost — togethercompute · 2026-07-27
- OpenAI may be hitting compute limits as Codex and ChatGPT Work jump from 2M to 10M users — JoshuaJBouw · 2026-07-27
- NVIDIA says Vera CPU is speeding up next-gen CPU and GPU design cycles — nordicinst · 2026-07-27
- NVIDIA says Nemotron 3 Ultra hit 97.1% on agentic RTL chip-design tasks — NVIDIAAI · 2026-07-27
- NVIDIA says Vera CPU lifted selected EDA workloads by up to 1.5x — NVIDIA Blog · 2026-07-27
- Local Qwen models power a robot that tests 78 smartphones’ battery life — gappyvalley · 2026-07-27