English translation is being prepared.
Matched fieldsSummary、Body
←Search results
Search “DeepSeek”
English translation is being prepared.
Matched fieldsSummary、Body
English translation is being prepared.
Matched fieldsSummary
English translation is being prepared.
Matched fieldsSummary、Body
DeepSeek open-sources the inference engine DeepInfer on GitHub
On October 29, 2026, DeepSeek open-sourced the inference engine DeepInfer on GitHub; this engine is the inference engine behind DeepSeek-V4.1-Flash. Leiphone reported the same day that DeepInfer includes technologies such as MoE expert parallelism, asynchronous batch scheduling, custom kernels, and speculative decoding, and disclosed that a single cluster needs to process approximately 155 trillion tokens per day. Its Zero-Overhead Batch Scheduler lets the CPU schedule computation one round in advance, overlapping with GPU forward passes, and handles cross-round state misalignment through total_inflight_tokens accounting, Remap mapping, and DeferredRelease deferred reclamation. Currently, only offline inference capabilities are open; components such as online KV Cache storage and cross-replica load balancing have not yet been fully disclosed, and the official team has not separately published end-to-end speedup data for this scheduling mechanism relative to synchronous solutions.
On October 29, 2026, DeepSeek open-sourced the inference engine DeepInfer on GitHub; this engine is the inference engine behind DeepSeek-V4.1-Flash. Leiphone reported the same day that DeepInfer includes technologies such as MoE expert parallelism, asynchronous batch scheduling, custom kernels, and speculative decoding, and disclosed that a single cluster needs to process approximately 155 trillion tokens per day. Its Zero-Overhead Batch Scheduler lets the CPU schedule computation one round in a
Matched fieldsTitle
vLLM Blog Publishes Technical Retrospective on Kimi K3 Inference Optimization
On October 16, 2026, the vLLM blog published a technical retrospective titled "Hyper Optimizing Kimi K3 in vLLM: Moving Fast and Responsibly." According to a TMTPost·AGI report, NVIDIA CentML engineer Benjamin Chislett used a Megakernel written in 12,000 lines of CuTe DSL to optimize inference for Moonshot AI's Kimi K3; the vLLM blog reports over 840 tokens/s at batch 1 and a 1.5x to 2x improvement in decoding throughput across multiple configurations, while SemiAnalysis estimates 423 tok/s/user on GB300 NVL72, a 1.68x improvement. The vLLM blog states that agent-optimized inference code is now outpacing human review and maintenance capacity, and the team therefore proposed a fast-track model optimization RFC so that reviewers judge designs from high-level abstractions rather than reviewing code line by line. The report also mentions that NVIDIA is developing optimizations for other models in vLLM, such as DeepSeek V4.1.
ltiple configurations, while SemiAnalysis estimates 423 tok/s/user on GB300 NVL72, a 1.68x improvement. The vLLM blog states that agent-optimized inference code is now outpacing human review and maintenance capacity, and the team therefore proposed a fast-track model optimization RFC so that reviewers judge designs from high-level abstractions rather than reviewing code line by line. The report also mentions that NVIDIA is developing optimizations for other models in vLLM, such as DeepSeek V4.1.
Matched fieldsSummary
DeepSeek open-sources inference framework DeepInfer
On October 20, 2026, DeepSeek open-sourced its internal inference framework DeepInfer. According to Yuchen Jin, citing data disclosed by DeepSeek, a single cluster processes 155 trillion tokens per day and generates $38 in revenue per GPU-hour (off-peak pricing); DeepInfer supports both NVIDIA Blackwell and Huawei Ascend. A subsequent 36Kr report stated that its accompanying open-source deepseek-recipe supports NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs. On October 21, swyx, citing Latent.Space data, said a single cluster processes 155T tokens per day with a 96% input cache hit rate, and argued that the lifetime value of each sold GPU has risen by about 30%. On October 24, Nathan Lambert said DeepSeek's single cluster processes 150T tokens per day, exceeding the combined open-source model inference volume of OpenRouter, Fireworks, Together, and BaseTen, and noted that the actual number of clusters is unknown; on the same day, LMSYS published a same-hardware benchmark comparison, stating that SGLang meets or exceeds DeepInfer on DeepInfer's own benchmark, with prefill throughput up to 1.04x and decode step length shortened by 1.1%–7.4% on 64x GB300.
On October 20, 2026, DeepSeek open-sourced its internal inference framework DeepInfer. According to Yuchen Jin, citing data disclosed by DeepSeek, a single cluster processes 155 trillion tokens per day and generates $38 in revenue per GPU-hour (off-peak pricing); DeepInfer supports both NVIDIA Blackwell and Huawei Ascend. A subsequent 36Kr report stated that its accompanying open-source deepseek-recipe supports NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs. On October 21, swyx, citing Latent.Spac
Matched fieldsTitle
Nathan Lambert says DeepSeek processes 150T tokens per day on a single cluster
On October 20, 2026, DeepSeek open-sourced its internal inference framework DeepInfer. According to Yuchen Jin, citing data disclosed by DeepSeek, a single cluster processes 155 trillion tokens per day and generates $38 in revenue per GPU-hour (off-peak pricing); DeepInfer supports both NVIDIA Blackwell and Huawei Ascend. A subsequent 36Kr report stated that its accompanying open-source deepseek-recipe supports NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs. On October 21, swyx, citing Latent.Space data, said a single cluster processes 155T tokens per day with a 96% input cache hit rate, and argued that the lifetime value of each sold GPU has risen by about 30%. On October 24, Nathan Lambert said DeepSeek's single cluster processes 150T tokens per day, exceeding the combined open-source model inference volume of OpenRouter, Fireworks, Together, and BaseTen, and noted that the actual number of clusters is unknown; on the same day, LMSYS published a same-hardware benchmark comparison, stating that SGLang meets or exceeds DeepInfer on DeepInfer's own benchmark, with prefill throughput up to 1.04x and decode step length shortened by 1.1%–7.4% on 64x GB300.
Matched fieldsTitle
View story:DeepSeek open-sources inference framework DeepInfer
DeepSeek integrates Harness with the web, App, and open platform teams
On October 16, 2026, DeepSeek established a product department, integrating its four teams—Harness, Web, App, and Platform—under the leadership of Cui Tianyi, who comes from a quantitative trading background, to oversee product, design, and R&D. Related reports noted that DeepSeek Harness was open-sourced under the MIT license on August 13, and by mid-October the repository had accumulated about 250,000 GitHub stars; the official website has posted job openings for two directions: AI product manager and Agent Harness. The reports also mentioned that Guo Daya, a core contributor to R1, left in March 2026 to join ByteDance Seed because he believed the Agent direction was not a high priority within DeepSeek. Currently, this integration is coordinated by Cui Tianyi, and there are no further developments with supporting evidence at this time.
On October 16, 2026, DeepSeek established a product department, integrating its four teams—Harness, Web, App, and Platform—under the leadership of Cui Tianyi, who comes from a quantitative trading background, to oversee product, design, and R&D. Related reports noted that DeepSeek Harness was open-sourced under the MIT license on August 13, and by mid-October the repository had accumulated about 250,000 GitHub stars; the official website has posted job openings for two directions: AI product man
Matched fieldsTitle
DeepSeek integrates the Harness team with the web, App, and open platform teams
On October 16, 2026, DeepSeek established a product department, integrating its four teams—Harness, Web, App, and Platform—under the leadership of Cui Tianyi, who comes from a quantitative trading background, to oversee product, design, and R&D. Related reports noted that DeepSeek Harness was open-sourced under the MIT license on August 13, and by mid-October the repository had accumulated about 250,000 GitHub stars; the official website has posted job openings for two directions: AI product manager and Agent Harness. The reports also mentioned that Guo Daya, a core contributor to R1, left in March 2026 to join ByteDance Seed because he believed the Agent direction was not a high priority within DeepSeek. Currently, this integration is coordinated by Cui Tianyi, and there are no further developments with supporting evidence at this time.
Matched fieldsTitle
View story:DeepSeek integrates Harness with the web, App, and open platform teams
DeepSeek open-sources inference framework DeepInfer
On October 20, 2026, DeepSeek announced the open-sourcing of its internal production-grade inference framework, DeepInfer. According to Latent Space, the framework powers DeepSeek-V4.1-Flash, is customized based on vLLM, supports distributed inference, speculative decoding, SWA Bounded Replay, and structured output, and targets the NVIDIA SM100 and Huawei Ascend 950 platforms; the published single-cluster production figures are 155 trillion tokens processed per day, a 96% input cache hit rate, 400 ms first-token latency, and a generation speed of 230 tokens/s. The report also noted that components such as KV cache management and load balancing have not yet been fully open-sourced. The next day, DeepSeek published the DeepInfer technical report on Zhihu, disclosing internal details including the hardware selection for Prefill and Decode and the KVCache targets, and providing data showing a single cluster serving 155 trillion Tokens/day at $38 per card per hour at off-peak prices. The framework is currently open-sourced, but the KV cache system, Prefill/Decode disaggregation, and load balancing required for a complete service are not fully included in this open-source project.
On October 20, 2026, DeepSeek announced the open-sourcing of its internal production-grade inference framework, DeepInfer. According to Latent Space, the framework powers DeepSeek-V4.1-Flash, is customized based on vLLM, supports distributed inference, speculative decoding, SWA Bounded Replay, and structured output, and targets the NVIDIA SM100 and Huawei Ascend 950 platforms; the published single-cluster production figures are 155 trillion tokens processed per day, a 96% input cache hit rate, 4
Matched fieldsTitle
DeepSeek open-sources inference framework DeepInfer
On October 20, 2026, DeepSeek open-sourced its internal inference framework DeepInfer. According to Yuchen Jin, citing data disclosed by DeepSeek, a single cluster processes 155 trillion tokens per day and generates $38 in revenue per GPU-hour (off-peak pricing); DeepInfer supports both NVIDIA Blackwell and Huawei Ascend. A subsequent 36Kr report stated that its accompanying open-source deepseek-recipe supports NVIDIA SM100 GPUs and Huawei Ascend 950 NPUs. On October 21, swyx, citing Latent.Space data, said a single cluster processes 155T tokens per day with a 96% input cache hit rate, and argued that the lifetime value of each sold GPU has risen by about 30%. On October 24, Nathan Lambert said DeepSeek's single cluster processes 150T tokens per day, exceeding the combined open-source model inference volume of OpenRouter, Fireworks, Together, and BaseTen, and noted that the actual number of clusters is unknown; on the same day, LMSYS published a same-hardware benchmark comparison, stating that SGLang meets or exceeds DeepInfer on DeepInfer's own benchmark, with prefill throughput up to 1.04x and decode step length shortened by 1.1%–7.4% on 64x GB300.
Matched fieldsTitle
View story:DeepSeek open-sources inference framework DeepInfer
DeepSeek releases DeepSeek-V4.1-Flash-1017
On October 17, 2026, DeepSeek released DeepSeek-V4.1-Flash-1017, which the company says delivers stronger benchmark performance, faster task completion, and shorter outputs, achieving a Pareto improvement in performance and cost; it is now available on the official website, App, and API, and the model also launched on OpenRouter the same day. The company also announced that the Search and Fetch APIs are open to everyone, priced at $0.75 per 1,000 calls per endpoint, with calls made through server-side tools in DeepSeek Harness, Claude Code, and Codex temporarily free, and billing starting at 16:00 UTC on October 31, 2026. On October 17, Thomas Wolf commented that DeepSeek called this a "minor update," yet Terminal-Bench 4.0 rose from 31.2 to 42.3. On October 19, Arena announced that the model, with a score of 1670, became the highest-ranked open-source model on Code Arena: WebDev. On October 20, OpenCode updated it to version 1017 in Go and Free; Unsloth released a GGUF that can run locally, with the 1-bit quantized version retaining about 88.2% accuracy and shrinking in size from 508GB to 250GB; Arena then announced that it topped the Agent Arena open-source model leaderboard.
On October 17, 2026, DeepSeek released DeepSeek-V4.1-Flash-1017, which the company says delivers stronger benchmark performance, faster task completion, and shorter outputs, achieving a Pareto improvement in performance and cost; it is now available on the official website, App, and API, and the model also launched on OpenRouter the same day. The company also announced that the Search and Fetch APIs are open to everyone, priced at $0.75 per 1,000 calls per endpoint, with calls made through serve
Matched fieldsTitle
DeepSeek releases DeepSeek-V4.1-Flash-1017
On October 17, 2026, DeepSeek released DeepSeek-V4.1-Flash-1017, which the company says delivers stronger benchmark performance, faster task completion, and shorter outputs, achieving a Pareto improvement in performance and cost; it is now available on the official website, App, and API, and the model also launched on OpenRouter the same day. The company also announced that the Search and Fetch APIs are open to everyone, priced at $0.75 per 1,000 calls per endpoint, with calls made through server-side tools in DeepSeek Harness, Claude Code, and Codex temporarily free, and billing starting at 16:00 UTC on October 31, 2026. On October 17, Thomas Wolf commented that DeepSeek called this a "minor update," yet Terminal-Bench 4.0 rose from 31.2 to 42.3. On October 19, Arena announced that the model, with a score of 1670, became the highest-ranked open-source model on Code Arena: WebDev. On October 20, OpenCode updated it to version 1017 in Go and Free; Unsloth released a GGUF that can run locally, with the 1-bit quantized version retaining about 88.2% accuracy and shrinking in size from 508GB to 250GB; Arena then announced that it topped the Agent Arena open-source model leaderboard.
Matched fieldsTitle
English translation is being prepared.
On October 17, 2026, DeepSeek released DeepSeek-V4.1-Flash-1017, which the company says delivers stronger benchmark performance, faster task completion, and shorter outputs, achieving a Pareto improvement in performance and cost; it is now available on the official website, App, and API, and the model also launched on OpenRouter the same day. The company also announced that the Search and Fetch APIs are open to everyone, priced at $0.75 per 1,000 calls per endpoint, with calls made through server-side tools in DeepSeek Harness, Claude Code, and Codex temporarily free, and billing starting at 16:00 UTC on October 31, 2026. On October 17, Thomas Wolf commented that DeepSeek called this a "minor update," yet Terminal-Bench 4.0 rose from 31.2 to 42.3. On October 19, Arena announced that the model, with a score of 1670, became the highest-ranked open-source model on Code Arena: WebDev. On October 20, OpenCode updated it to version 1017 in Go and Free; Unsloth released a GGUF that can run locally, with the 1-bit quantized version retaining about 88.2% accuracy and shrinking in size from 508GB to 250GB; Arena then announced that it topped the Agent Arena open-source model leaderboard.
Matched fieldsTitle
English translation is being prepared.
Matched fieldsTitle
English translation is being prepared.
Matched fieldsTitle
China's Ministry of Commerce and Ministry of Science and Technology plan to expand the export control technology catalog
On October 16, 2026, China's Ministry of Commerce and Ministry of Science and Technology issued Announcement No. 48 of 2026, soliciting public comments on revisions to the Catalog of Technologies Prohibited or Restricted from Export in China, proposing to add 40 new entries, of which 14 are prohibited from export and 26 are restricted from export, and effective immediately, all newly added technologies are subject to pre-export registration. The draft for the first time includes advanced large models, AI agents, and large-scale computing clusters, setting three thresholds for frontier models: training compute of no less than 10^26 floating-point operations, no fewer than 2 trillion parameters, or reaching a globally leading level; the notes specify that technologies that have already entered the public domain through open-source or academic publication are not within the scope of control, and that open-source releases, enterprises building and operating their own overseas computing centers, and similar activities may obtain automatic licenses. The comment period runs until November 15. On October 23, DeepSeek and MiniMax completed registration for open-source and overseas authorized use, and will continue to open-source their models.
ecify that technologies that have already entered the public domain through open-source or academic publication are not within the scope of control, and that open-source releases, enterprises building and operating their own overseas computing centers, and similar activities may obtain automatic licenses. The comment period runs until November 15. On October 23, DeepSeek and MiniMax completed registration for open-source and overseas authorized use, and will continue to open-source their models.
Matched fieldsSummary
Matt Turck shares his conversation with Pierre Stock about Le Chonk
Matt Turck shared his conversation with Mistral research lead Pierre Stock about Le Chonk. According to the conversation, Le Chonk is described as the largest open-source model in the West, trained by the team on 3,800 GPUs in their own data center, and is still in training. Stock also discussed what Mistral borrowed from DeepSeek, and the trade-off between open-source weights and cybersecurity refusals. The program also mentioned that Mistral Large 5 will be released later.
Matched fieldsSummary