# V4.1-Flash

Published articles for V4.1-Flash.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1

DevFeed: [Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1](<https://devfeed.tech/articles/zero-mismatch-simple-async-rl-reinforce-zero-deepseek-v4-1-and-kimi-k3-1-122091.md>)

Original publisher: [Read original article](<https://vllm.ai/blog/2026-11-06-zero-mismatch-async-rl>)

Author: Ao Shen (Inferact), Jian Hu (NVIDIA Research), Yan Bai (NVIDIA DevTech), Shekhar Pandey (AMD), and the REINFORCE Zero authors

Published: 2026-11-06T00:00:00Z

Content type: article

Language: en

Sources: [vLLM Blog](<https://devfeed.tech/sources/vllm-blog.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [async](<https://devfeed.tech/topics/async.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [quantization](<https://devfeed.tech/topics/quantization.md>), [grpo](<https://devfeed.tech/topics/grpo.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [3](<https://devfeed.tech/tags/3.md>), [7-4](<https://devfeed.tech/tags/7-4.md>), [8k](<https://devfeed.tech/tags/8k.md>), [about](<https://devfeed.tech/tags/about.md>), [ai-engineering](<https://devfeed.tech/tags/ai-engineering.md>), [align](<https://devfeed.tech/tags/align.md>), [amd](<https://devfeed.tech/tags/amd.md>), [asynchronous](<https://devfeed.tech/tags/asynchronous.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepseek-v4-1-flash](<https://devfeed.tech/tags/deepseek-v4-1-flash.md>), [flash](<https://devfeed.tech/tags/flash.md>), [kernels](<https://devfeed.tech/tags/kernels.md>), [kimi](<https://devfeed.tech/tags/kimi.md>), [lag](<https://devfeed.tech/tags/lag.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [model-serving](<https://devfeed.tech/tags/model-serving.md>), [open-source-ai](<https://devfeed.tech/tags/open-source-ai.md>), [reinforcement-learning](<https://devfeed.tech/tags/reinforcement-learning.md>), [rl](<https://devfeed.tech/tags/rl.md>), [sampling](<https://devfeed.tech/tags/sampling.md>), [training](<https://devfeed.tech/tags/training.md>), [trains](<https://devfeed.tech/tags/trains.md>), [v4-1-flash](<https://devfeed.tech/tags/v4-1-flash.md>), [vllm-blog](<https://devfeed.tech/tags/vllm-blog.md>), [zero](<https://devfeed.tech/tags/zero.md>)

### AI overview

REINFORCE Zero argues that asynchronous RL can train stably on stale samples when rollout and learner kernels are aligned, leaving unclipped token level importance sampling to correct weight lag. The paper reports stable training through 20,000 updates and learning at policy lag 64; in one comparison, importance weight deviation peaked at 1.50 times its median with aligned kernels versus 442.54 times with standard kernels. Without alignment, training drifted and a sequence gate rejected 24.48% of sequences. Alignment has a speed cost: one OSWorld setup took 6.8% longer per step than standard kernel IS with a gate. The excerpt also reports zero-mismatch deployments for DeepSeek-V4.1-Flash and Kimi K3.1, including AMD MI355X runs, while noting ongoing performance work and that long asynchronous runs on the flagship models remain future work.

### Source excerpt

Asynchronous RL trains on stale weights. Align rollout and training kernels bit for bit, and unclipped importance sampling corrects that lag. Plus zero-mismatch RL for DeepSeek-V4.1-Flash, Kimi K3.1 and AMD MI355X.

## Serving DeepSeek-V4.1-Flash at Scale: vLLM Matches DeepInfer and Scales to GB300 NVL72

DevFeed: [Serving DeepSeek-V4.1-Flash at Scale: vLLM Matches DeepInfer and Scales to GB300 NVL72](<https://devfeed.tech/articles/serving-deepseek-v4-1-flash-at-scale-vllm-matches-deepinfer-and-scales-to-gb300-nvl72-98978.md>)

Original publisher: [Read original article](<https://vllm.ai/blog/2026-10-23-deepseek-v41-flash-pd-wide-ep>)

Author: Inferact and the vLLM Team

Published: 2026-10-23T00:00:00Z

Content type: article

Language: en

Sources: [vLLM Blog](<https://devfeed.tech/sources/vllm-blog.md>)

Topics: [vllm](<https://devfeed.tech/topics/vllm.md>), [GB300 NVL72](<https://devfeed.tech/topics/gb300-nvl72.md>), [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [LLM Techniques](<https://devfeed.tech/topics/llm-techniques.md>), [NVLink](<https://devfeed.tech/topics/nvlink.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [ai-engineering](<https://devfeed.tech/tags/ai-engineering.md>), [async](<https://devfeed.tech/tags/async.md>), [attention](<https://devfeed.tech/tags/attention.md>), [b300](<https://devfeed.tech/tags/b300.md>), [balancing](<https://devfeed.tech/tags/balancing.md>), [deepinfer](<https://devfeed.tech/tags/deepinfer.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepseek-v4-1-flash](<https://devfeed.tech/tags/deepseek-v4-1-flash.md>), [disaggregation](<https://devfeed.tech/tags/disaggregation.md>), [gb300](<https://devfeed.tech/tags/gb300.md>), [gb300-nvl72](<https://devfeed.tech/tags/gb300-nvl72.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [large-scale-serving](<https://devfeed.tech/tags/large-scale-serving.md>), [llm-inference](<https://devfeed.tech/tags/llm-inference.md>), [model-serving](<https://devfeed.tech/tags/model-serving.md>), [nvl72](<https://devfeed.tech/tags/nvl72.md>), [open-source-ai](<https://devfeed.tech/tags/open-source-ai.md>), [performance](<https://devfeed.tech/tags/performance.md>), [rack](<https://devfeed.tech/tags/rack.md>), [scale](<https://devfeed.tech/tags/scale.md>), [tokens](<https://devfeed.tech/tags/tokens.md>), [user](<https://devfeed.tech/tags/user.md>), [v4-1-flash](<https://devfeed.tech/tags/v4-1-flash.md>), [vllm](<https://devfeed.tech/tags/vllm.md>), [vllm-blog](<https://devfeed.tech/tags/vllm-blog.md>), [wide](<https://devfeed.tech/tags/wide.md>)

### AI overview

On one GB300 NVL72 rack, vLLM's P/D-disaggregated, Wide-EP serving of DeepSeek-V4.1-Flash reaches 954K tokens/s per GPU on SemiAnalysis AgentX at a P90 of 196 tokens/s per user, reported as 7.9x the authors' earlier single-node result. Decode throughput on GB200 is within 6% of DeepSeek's DeepInfer across batch sizes 8-256; the prefill curves track, though differing setups prevent a direct speedup comparison. The reported gains come from fitting prefill to DEP2, extending bounded replay to layer 20, using larger GPU pages for shared Engram tables and DeepSelect for prefill ranking, plus Wide-EP with EPLB for decode. At DEP32, EPLB substantially improves expert-load balance, while CPU overhead can erase GPU-side scaling gains. A specialized decode path reaches 826 P90 tokens/s for one user on four B300 GPUs, 44% above the standard path. Results depend on the stated hardware, configurations and workloads.

### Source excerpt

With P/D disaggregation and Wide-EP on a GB300 NVL72 rack, vLLM serves DeepSeek-V4.1-Flash on SemiAnalysis AgentX at 954K tokens/s per GPU at a P90 of 196 tokens/s per user, 7.9x our aggregated single-node result.

## MiniMax launches M3.1 Flash model, says it cuts AI agent costs by 70%

DevFeed: [MiniMax launches M3.1 Flash model, says it cuts AI agent costs by 70%](<https://devfeed.tech/articles/minimax-launches-m3-1-flash-model-says-it-cuts-ai-agent-costs-by-70-90485.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/19/minimax-launches-m3-1-flash-model-says-it-cuts-ai-agent-costs-by-70/>)

Author: Duncan Riley

Published: 2026-10-19T22:25:50Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [LLM optimization](<https://devfeed.tech/topics/llm-optimization.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [LLM evaluation / benchmarking](<https://devfeed.tech/topics/llm-evaluation-benchmarking.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-costs](<https://devfeed.tech/tags/agent-costs.md>), [agentic-tasks](<https://devfeed.tech/tags/agentic-tasks.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [api](<https://devfeed.tech/tags/api.md>), [bankertoolbench](<https://devfeed.tech/tags/bankertoolbench.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [berkeley](<https://devfeed.tech/tags/berkeley.md>), [cybergym](<https://devfeed.tech/tags/cybergym.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepswe](<https://devfeed.tech/tags/deepswe.md>), [hangzhou-deepseek](<https://devfeed.tech/tags/hangzhou-deepseek.md>), [hong-kong](<https://devfeed.tech/tags/hong-kong.md>), [inference](<https://devfeed.tech/tags/inference.md>), [initial-public-offering](<https://devfeed.tech/tags/initial-public-offering.md>), [large-language-model](<https://devfeed.tech/tags/large-language-model.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [m-plan](<https://devfeed.tech/tags/m-plan.md>), [m3](<https://devfeed.tech/tags/m3.md>), [m3-1-flash](<https://devfeed.tech/tags/m3-1-flash.md>), [minimax](<https://devfeed.tech/tags/minimax.md>), [minimax-code](<https://devfeed.tech/tags/minimax-code.md>), [model-weights](<https://devfeed.tech/tags/model-weights.md>), [news](<https://devfeed.tech/tags/news.md>), [openrouter](<https://devfeed.tech/tags/openrouter.md>), [osworld-v2](<https://devfeed.tech/tags/osworld-v2.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [release](<https://devfeed.tech/tags/release.md>), [space-bunny](<https://devfeed.tech/tags/space-bunny.md>), [sparsity](<https://devfeed.tech/tags/sparsity.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-3](<https://devfeed.tech/tags/top-story-3.md>), [top-story-6](<https://devfeed.tech/tags/top-story-6.md>), [university-of-california](<https://devfeed.tech/tags/university-of-california.md>), [v4-1-flash](<https://devfeed.tech/tags/v4-1-flash.md>)

### AI overview

MiniMax released M3.1 Flash, a faster, lower-cost update to its M3 large language model aimed at reducing the cost of running AI agents. The company reports 70% lower agentic-task costs and inference speeds close to three times faster than its previous model, along with benchmark results for computer operation, banking workflows, cybersecurity, and software engineering. The model is available through MiniMax's API, subscription plan, and desktop app; the company says it plans to release the weights later.

### Source excerpt

Chinese artificial intelligence developer MiniMax Group Inc. today released M3.1 Flash, a faster and cheaper update to the M3 large language model it launched in June. The release takes aim at the cost of running AI agents, which often need dozens of model calls to finish a single job. MiniMax said Flash represents its "ongoing [...] The post MiniMax launches M3.1 Flash model, says it cuts AI agent costs by 70% appeared first on SiliconANGLE.

## DeepSeek releases updated V4.1-Flash model it says completes tasks faster

DevFeed: [DeepSeek releases updated V4.1-Flash model it says completes tasks faster](<https://devfeed.tech/articles/deepseek-releases-updated-v4-1-flash-model-it-says-completes-tasks-faster-89813.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/18/deepseek-releases-updated-v4-1-flash-model-it-says-completes-tasks-faster/>)

Author: Duncan Riley

Published: 2026-10-18T22:34:04Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [Large Language Model](<https://devfeed.tech/topics/llm.md>), [releases](<https://devfeed.tech/topics/releases.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [agent-software](<https://devfeed.tech/tags/agent-software.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [alibaba](<https://devfeed.tech/tags/alibaba.md>), [alibaba-group](<https://devfeed.tech/tags/alibaba-group.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [anthropic-pbc](<https://devfeed.tech/tags/anthropic-pbc.md>), [api](<https://devfeed.tech/tags/api.md>), [application-programming-interface](<https://devfeed.tech/tags/application-programming-interface.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [claude-opus-5](<https://devfeed.tech/tags/claude-opus-5.md>), [codebuddy](<https://devfeed.tech/tags/codebuddy.md>), [codex](<https://devfeed.tech/tags/codex.md>), [deepseek](<https://devfeed.tech/tags/deepseek.md>), [deepseek-harness](<https://devfeed.tech/tags/deepseek-harness.md>), [deepseek-v4-1-flash](<https://devfeed.tech/tags/deepseek-v4-1-flash.md>), [deepseek-v4-1-flash-1017](<https://devfeed.tech/tags/deepseek-v4-1-flash-1017.md>), [deepswe](<https://devfeed.tech/tags/deepswe.md>), [fetch-api](<https://devfeed.tech/tags/fetch-api.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [humanitys-last-exam](<https://devfeed.tech/tags/humanitys-last-exam.md>), [minimax](<https://devfeed.tech/tags/minimax.md>), [minimax-group](<https://devfeed.tech/tags/minimax-group.md>), [model](<https://devfeed.tech/tags/model.md>), [news](<https://devfeed.tech/tags/news.md>), [openai](<https://devfeed.tech/tags/openai.md>), [original](<https://devfeed.tech/tags/original.md>), [outage](<https://devfeed.tech/tags/outage.md>), [releases](<https://devfeed.tech/tags/releases.md>), [search-api](<https://devfeed.tech/tags/search-api.md>), [siliconangle](<https://devfeed.tech/tags/siliconangle.md>), [tasks](<https://devfeed.tech/tags/tasks.md>), [tencent](<https://devfeed.tech/tags/tencent.md>), [tencent-holdings](<https://devfeed.tech/tags/tencent-holdings.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-6](<https://devfeed.tech/tags/top-story-6.md>), [v4-1-flash](<https://devfeed.tech/tags/v4-1-flash.md>), [workbuddy](<https://devfeed.tech/tags/workbuddy.md>)

### AI overview

DeepSeek released DeepSeek-V4.1-Flash-1017, an update it says reduces task execution time and output length. The model scored 77% on DeepSeek's DeepSWE v1.1 software engineering test, compared with 74.2% for the prior version; the article also reports that seven of 11 published benchmarks favored Claude Opus 5. The update is tuned for agent software, and DeepSeek also opened Search and Fetch APIs to developers. The model is available through the existing API model ID at unchanged prices, with weights released under the MIT license.

### Source excerpt

Chinese artificial intelligence startup Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co. Ltd. has released an updated version of its V4.1-Flash model that it says completes tasks faster. DeepSeek-V4.1-Flash-1017 arrives a little over five weeks after the company launched the original V4.1-Flash as the smallest model in a new architecture family. Its model card on [...] The post DeepSeek releases updated V4.1-Flash model it says completes tasks faster appeared first on SiliconANGLE.