Observed Signal · Oct 20, 2026 · Technical Release · Source: Trending Topics (DACH/CEE Innovation & Tech) · Impact: 3/5 · Sentiment: Neutral
DeepSeek Open-Sources DeepInfer Inference Engine
DeepSeek open-sourced DeepInfer, the production inference engine behind its V4.1 Flash model, under Apache 2.0 during PyTorch Conference 2026. The framework supports Nvidia Blackwell (including SM100) and Huawei Ascend 950 chips, but not older Hopper, aligning with DeepSeek's strategy to reduce Nvidia dependence. It integrates previously open-sourced components like FlashMLA, DeepGEMM, and DeepEP, and includes techniques for distributed inference and speculative decoding. However, key features such as cache management, distributed execution, and load balancing remain proprietary. DeepSeek disclosed impressive metrics: 155 trillion tokens per day per cluster, 96% input cache hits, and $38 per accelerator-hour at off-peak, suggesting high margins. vLLM claims its throughput is within 6% of DeepInfer's, sparking debate. The release coincides with U.S. government actions against DeepSeek.
DeepSeek's open-sourcing of DeepInfer is significant for the AI infrastructure landscape, showcasing alternative hardware support (Huawei) and reducing Nvidia dependence. However, the incomplete release limits immediate industry impact.
Track DeepSeek Signals & Market Shifts in Real-Time
Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your free Explorer workspace to monitor these entities.
Key Takeaways & Evidence Grounding
- DeepSeek released DeepInfer as open source under Apache 2.0.
- DeepInfer supports Nvidia Blackwell and Huawei Ascend 950, but not older Hopper.
- DeepSeek reports 155 trillion tokens per day per cluster and $38 revenue per accelerator-hour at off-peak.
- Key production features like cache management, distributed execution, and load balancing remain proprietary.
- vLLM claims decode throughput within 6% of DeepInfer's results.
Connected Companies & Entities
5 Entities mapped“Chinese AI lab that released DeepInfer as open source.”
“Hardware vendor whose Ascend 950 chips are supported by DeepInfer.”
“Hardware vendor whose Blackwell chips are supported by DeepInfer.”
“The vLLM team reported decode throughput within 6% of DeepInfer's published results.”
“OpenAI announced 12-hour API prompt caching, compared to DeepSeek's 72 hours.”
Ontology Mapping & Concepts
Related Market Signals & Shifts
Recent verified developments and strategic activity across this market segment.
DeepSeek Founder: Compute Is the Primary Constraint
Hello China Tech published a July 2026 selection from a nearly four-hour investor transcript in which DeepSeek founder Liang Wenfeng argued that compute availability is the principal gap between DeepSeek and leading US AI labs. Liang framed differences in talent, model capability, and applications as downstream consequences of smaller compute budgets and limited chip supply. He placed DeepSeek’s first external round at over Rmb 50bn (~$7.4bn) (not officially confirmed by the company), described work to reduce dependence on Nvidia via a high-level compiler called TileLang, and forecast that within a year domestic Chinese chips could be verified as usable for training. Liang also discussed pricing, open-weight releases, team retention via option grants, and V4 multimodality commitments.
DeepSeek V4 (MODEL1) Expected with Engram, mHC
DeepSeek, a Chinese open-source AI startup, is expected to release DeepSeek V4 (rumored codename MODEL1) around the Lunar New Year (week of Feb 17, 2026). Reporting and code commits indicate V4 will be a major architectural overhaul focused on extreme long-context coding and software-engineering tasks. Key innovations described include Engram (a conditional memory lookup to separate factual recall from reasoning and enable multi-million-token knowledge stores), Manifold-Constrained Hyper-Connections (mHC) to stabilize rich cross-layer connectivity, and DeepSeek Sparse Attention (DSA) for 1M+ token contexts. DeepSeek reportedly delayed its R2 training after hardware instability with Huawei Ascend chips and reverted to Nvidia GPUs for final training. The article places DeepSeek within a broader surge of Chinese open-weight model activity (names cited include Qwen/Alibaba Cloud, Zhipu AI, Moonshot AI, and Minimax).
DeepSeek Enters U.S. Corporate AI Spending
Chinese AI model provider DeepSeek has begun appearing in U.S. corporate vendor payments, according to Ramp’s June 2026 trending vendors list. The move follows DeepSeek’s May decision to make a 75% price cut on its V4‑Pro model permanent, setting cached-input pricing at RMB 0.025 (about $0.0035) per million tokens — roughly 1% of comparable cached-input costs cited for Anthropic. Ramp’s trending data contrasts with earlier developer-side signals from OpenRouter, which in May showed Chinese-built models dominating developer routing and weekly token volumes rising to ~25 trillion. Ramp cautioned that a monthly trending list is not conclusive market-share proof, but the appearance of other inference platforms (Fireworks AI, fal AI, DeepInfra) and GPU provider Vast.ai on the list suggests a growing cost-optimization infrastructure around alternatives to premium U.S. providers. Publication date: 2026-06-04.
Receive all future signals for DeepSeek & competitors in your briefing
Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.