Tagged articles

Ascend NPU

9 articles · Page 1 of 1

Sep 11, 2026 · Artificial Intelligence

openJiuwen Launches Dual-Dimensional RSI Framework for Self-Improving AI Agents

openJiuwen introduces a dual-dimensional Recursive Self-Improvement (RSI) framework that enables AI agents to automatically optimize both their tooling (Harness) and deliverables (research papers, algorithms) on the WorkSwarm platform, with compute-affinity scheduling on Ascend NPUs cutting latency and resource usage, validated by SWE-bench pass-rate gains from 61% to 87%.

AI agentsAscend NPUCompute Affinity

0 likes · 15 min read

openJiuwen Launches Dual-Dimensional RSI Framework for Self-Improving AI Agents

Huawei Cloud Developer Alliance

Sep 8, 2026 · Artificial Intelligence

AReaL v1.0.5 LoRA RL: Low-Rank Adaptation for Accessible Large Model RL on Ascend

This article details AReaL-Ascend v1.0.5's LoRA RL capabilities, explaining how low-rank adaptation reduces memory overhead for large model reinforcement learning, describing two Megatron LoRA weight update modes (adapter sync vs. merge), and covering cross-node LoRA RL, MoE support, XCCL communication, and Qwen3.6-27B examples for practical deployment.

AReaLAscend NPULoRA

0 likes · 6 min read

AReaL v1.0.5 LoRA RL: Low-Rank Adaptation for Accessible Large Model RL on Ascend

Huawei Cloud Developer Alliance

Sep 6, 2026 · Artificial Intelligence

AReaL v1.0.5: Colocated Training/Inference & Multi-Teacher On-Policy Distillation for Efficient RL

AReaL-Ascend v1.0.5 introduces two major innovations: colocated training and inference on shared NPUs via Ray scheduling and AWEX IPC zero-copy, and Multi-Teacher On-Policy Distillation (MOPD) that fuses domain expert models into a single student using token-level teacher signals, demonstrated on Qwen3.6-27B with GRPO.

AReaLAWEXAscend NPU

0 likes · 12 min read

AReaL v1.0.5: Colocated Training/Inference & Multi-Teacher On-Policy Distillation for Efficient RL

Huawei Cloud Developer Alliance

Sep 5, 2026 · Artificial Intelligence

THD Packing & MTP Speculative Decoding Tame 64K Code Agent RL on Ascend NPUs

AReaL-Ascend v1.0.5 enables stable 64K-context Code Agent RL for Qwen3.6-27B on Ascend NPUs, using THD Sequence Packing to eliminate padding waste and MTP speculative decoding to accelerate rollout, achieving reward improvement from 0.56 to 0.94 over 120 steps with train-rollout consistency near 1.0.

AReaLAscend NPUCode Agent

0 likes · 12 min read

THD Packing & MTP Speculative Decoding Tame 64K Code Agent RL on Ascend NPUs

Machine Heart

Aug 11, 2026 · Artificial Intelligence

openJiuwen and Ascend Enable Agent “Compute‑Affinity”: Halve First‑Token Latency, Cut Inference Storage by 25%

The openJiuwen platform introduces a semantic coordination layer called Agent Hint, together with SAM and SPM managers, to align agent task states with compute resources, achieving a 57% reduction in first‑token latency, a 27.6% drop in end‑to‑end latency, a 33% increase in cache hit rate, and a 25% decrease in storage peak for multi‑agent inference workloads.

Agent HintAscend NPUCompute Affinity

0 likes · 11 min read

openJiuwen and Ascend Enable Agent “Compute‑Affinity”: Halve First‑Token Latency, Cut Inference Storage by 25%

Machine Heart

Aug 8, 2026 · Artificial Intelligence

AURORA‑LM: Continuous‑Latent Diffusion Language Model on Ascend NPU

Researchers from Nanjing University’s PRLab present AURORA‑LM, a continuous‑latent diffusion language model trained on Ascend NPU up to 1 B parameters, detailing its two‑stage architecture, high‑capacity latent design, self‑trajectory consistency loss, and experimental gains over autoregressive and discrete diffusion baselines.

AURORA-LMAscend NPUcontinuous diffusion

0 likes · 10 min read

AURORA‑LM: Continuous‑Latent Diffusion Language Model on Ascend NPU

Huawei Cloud Developer Alliance

Apr 29, 2026 · Artificial Intelligence

Deploy DeepSeek‑V4 on Ascend NPU with Kthena in 3 Minutes (Prefill‑Decode Separation)

This guide walks through deploying the DeepSeek‑V4‑Flash model on Ascend NPU using Kthena’s ModelRoute, detailing the Prefill‑Decode (P/D) separation architecture, KV cache transfer via Mooncake, configuration of ModelServing and ModelRoute resources, and flexible scaling of Prefill and Decode replicas for optimal performance.

Ascend NPUDeepSeek-V4KV Cache

0 likes · 22 min read

Deploy DeepSeek‑V4 on Ascend NPU with Kthena in 3 Minutes (Prefill‑Decode Separation)

Huawei Cloud Developer Alliance

Apr 13, 2026 · Artificial Intelligence

How AReaL v1.0 Enables Scalable Agentic RL on Ascend NPU with AWEX Weight Sync

The new AReaL v1.0 release brings full Ascend NPU support, detailed installation guides, and a best‑practice example for training a 30B MoE model across four nodes, while the integrated AWEX weight‑sync mechanism dramatically reduces synchronization time, improving efficiency and stability for large‑scale Agentic RL workloads.

AWEXAgentic RLAscend NPU

0 likes · 12 min read

How AReaL v1.0 Enables Scalable Agentic RL on Ascend NPU with AWEX Weight Sync

AIWalker

Apr 13, 2025 · Artificial Intelligence

Huawei Pangu Ultra: 135B Ascend‑Native Dense LLM Without Nvidia GPUs

Huawei's Pangu Ultra introduces a 135‑billion‑parameter dense language model trained entirely on Ascend NPUs, detailing novel stability architectures, a domain‑aware tokenizer, multi‑stage pre‑training, extensive system optimizations, and benchmark results that surpass leading models such as Llama 405B and DeepSeek‑R1.

Ascend NPUDense ModelLarge Language Model

0 likes · 15 min read

Huawei Pangu Ultra: 135B Ascend‑Native Dense LLM Without Nvidia GPUs