Search: “FastH3”

31 stories

DomainAll Cyber AIWindow24h 3d 7d 30d AllIn the wild Important only Show duplicates

SeverityAll critical 0 high 0 medium 0 low 0 info 1

Smart search ranks by meaning as well as keywords (one row per story, last 45 days).

Sparks Fly: NVIDIA Accelerates Local AI at IFA 2026

NVIDIA announces local AI push at IFA 2026 with faster llama.cpp/vLLM inference, PAIR routing tool, and October RTX Spark PCs.

At IFA 2026, NVIDIA announced simplified local AI support for agents in Hermes Agent, OpenClaw, and Perplexity Portable Computer, plus new llama.cpp and vLLM optimizations delivering up to 1.9x faster local inference. NVIDIA also unveiled PAIR, a Personal AI Router for distributing inference across a local network's PCs, and compact RTX Spark Windows PCs from Lenovo and Acer arriving in October. The post recaps recent local-capable model releases including Nemotron 3.5 Lightning (30B), Qwen3.8-Flash-Next and Qwen3.8-27B, DeepSeek v4 Flash (284B MoE, 13B active), Meta Muse Glimmer (30B), Z.ai GLM-5.3-Flash, LTX 2.5, and MiniMax-H3 with the FastH3 distilled variant.

NVIDIA BlogSep 3, 16:00 UTC · 13d agoAI industry

WarmBloodAban/Minimax-h3_Singularity — new model trending #22 on Hugging Face

Community fine-tune Minimax-h3_Singularity enhances MiniMax-H3 video generation with HDR quality, distant face restoration, and improved motion, trending #22 on Hugging Face.

Minimax-h3_Singularity is a community fusion fine-tune of the MiniMax-H3 multimodal video generation model, built from multiple checkpoints and refined with pruning and weight optimization. It supports Text-to-Video, Image-to-Video, Reference-to-Video, and Video-to-Video workflows in ComfyUI, and claims improvements in HDR clarity, distant face restoration, motion fluidity, and fantasy VFX. The authors recommend pairing it with the minimax_h3_ref2v_turbo_4step_v0.1 LoRA for four-step accelerated inference, and an online demo is available via RunningHub.

Hugging Face trending modelsSep 5, 18:51 UTC · 11d agoModel release7· 1 read

[AINews] Fal’s H3 Max Live breaks the infinite videogen barrier

Fal post-trained MiniMax H3 into a 'Max' variant with 35x-faster inference, enabling faster-than-realtime AI video generation and infinite streams.

Fal post-trained MiniMax's H3 model into a 'Max' variant and optimized it for its in-house inference engine, achieving roughly 35x the speed of the official endpoint. The optimization enables faster-than-realtime video generation, demonstrated by an infinite interactive AI-generated stream productized by levels.io. The roundup also notes Meta Muse Code's general availability with an SDK, open DeepSeek-V4-Flash-Vision-Exp weights, GLM-5.3-Flash's strong agentic cost/performance rankings, and Tencent's 770B-parameter Hy4 Preview MoE with 49B active parameters.

Latent SpaceSep 1, 04:36 UTC · 15d agoAI industry

Jackrong/Qwopus3.8-27B-Flash-GGUF — new model trending #26 on Hugging Face

Community fine-tune Qwopus3.8-27B-Flash, built on Qwen3.8-27B, cuts agent reasoning latency with 12.8% faster decoding and 80.7% MTP acceptance.

Jackrong released Qwopus3.8-27B-Flash, a fine-tune of Qwen3.8-27B optimized for long-running agent workloads, reporting 12.8% faster decoding and 80.7% multi-token-prediction acceptance. Training used roughly 1.5 million teacher-scored SFT examples filtered to the top 10%, followed by reinforcement training with NVIDIA NeMo-RL and GSPO. The author notes an explicit trade-off: MMLU-Pro mixed-set scores are lower than the base model, and a known bug can produce incorrect Python indentation. Author-provided benchmarks have not been independently verified.

Hugging Face trending modelsSep 4, 12:43 UTC · 12d agoModel release1

OpenVDN/vdn-minimax-h3 — new model trending #12 on Hugging Face

OpenVDN releases VDN-H3, an open hybrid-attention video model on MiniMax H3 that renders a 14.4-second 768p clip in 11.23 seconds on 8 B200 GPUs.

VDN-Minimax-H3 (VDN-H3) adds a frame-wise linear attention branch plus two LoRA adapters to MiniMax H3, distilled into 8-step and 50-step variants. It generates 768p, 14.4-second clips in 11.23 seconds on 8 B200 GPUs (90.5 seconds on one H200) using 8 denoising steps. Weights (about 82 GB total, including the 72 GB H3 base), the optimized inference stack, and training code are fully open-source under the MiniMax H3 Community License, which excludes the EU, UK, Korea, and US.

Hugging Face trending modelsSep 2, 05:03 UTC · 14d agoModel release1

UniH^3: Unifying Hierarchical Homogeneity and Heterogeneity for All-in-One Medical Image Restoration

UniH3 unifies hierarchical homogeneity and heterogeneity modeling for all-in-one medical image restoration across modalities and degradation types.

UniH3 introduces a Hierarchical Homogeneity Memory module that distills shared anatomical priors from high-quality images, injected via a Homogeneity-Guided Attention mechanism. A Hierarchical Heterogeneity Balancer mitigates inter- and intra-task conflicts during multi-task optimization. It achieves state-of-the-art on MedIR-2D-500K and MedIR-3D-3D benchmarks for both all-in-one and single-task restoration, with code released on GitHub.

Hugging Face daily papersSep 9, 20:00 UTC · 6d agoAI research

m-a-p/YuE2-3B — new model trending #30 on Hugging Face

M-A-P released YuE2-3B, an open music generation model that outperforms Suno v5 on WildSongBench and runs locally on a 24GB GPU.

The M-A-P (multimodal-art-projection) team released YuE2-3B, an open-weights music generation model that turns lyrics and a style prompt into full songs with vocals and accompaniment. It uses an AR-NAR Mixture-of-Transformers backbone with symbolic planning and flow matching through a VAE, and supports editable scores (melody and chords, including ABC notation) plus agentic editing workflows. On 192 WildSongBench prompts it reports a SongBench average of 6.9632 (best-of-8) versus 6.8721 for Suno v5, claimed as state of the art among evaluated open and proprietary models. It runs 48 kHz stereo inference locally on a single 24GB NVIDIA GPU without quantization, with companion releases including YuE2-Vae, MERT-v2 encoders, the WildSongBench dataset, and SheetSage2.

Hugging Face trending modelsSep 9, 18:30 UTC · 7d agoModel release1

Kimi K3 (2.8T) at 1 token/s on a MacBook Pro, streamed from four SSDs

A GitHub project streams Moonshot's 2.8T-parameter Kimi K3 from four SSDs to run at 1 token/s on a MacBook Pro.

A GitHub repository (argonautlabsai/deltafin) demonstrates running Moonshot AI's Kimi K3, a 2.8-trillion-parameter model, on a MacBook Pro by streaming its weights from four SSDs at 1 token per second. The project drew 55 points and 22 comments on Hacker News, highlighting consumer-hardware inference for extremely large models.

Hacker News · AISep 8, 20:07 UTC · 7d agoAI tools & infra HN 55↑ · 22 comments

unsloth/Qwen3.8-Flash-Next-GGUF — new model trending #21 on Hugging Face

Qwen released Qwen3.8-Flash-Next, an experimental 125B-parameter open-weight MoE previewing the Qwen4 architecture, with Unsloth shipping optimized GGUF quants.

Qwen released Qwen3.8-Flash-Next, an experimental open-weight preview of the architecture planned to underpin Qwen4. The model has 125B parameters with 6B activated, 512 experts (10 routed plus 1 shared), Qwen Sparse Attention (QSA), Gated DeltaNet, Gated Residual, and n-gram embeddings, with 262,144-token native context extendable to 1,000,000 tokens. Unsloth provides Dynamic 3.0 GGUF quantizations, and multi-token prediction (MTP) delivers 1.3-1.7x faster inference via llama.cpp or Unsloth Desktop.

Hugging Face trending modelsAug 26, 12:29 UTC · 21d agoModel release1

I tested 10 model/harness combinations on the same Three.js task

A developer benchmarked 10 model/harness combinations on a Three.js task; Qwen 3.8 27B on OpenCode scored 95.64% fastest at 8m48s.

The author ran an identical Three.js sci-fi hangar build prompt across 10 model/harness combinations and recorded score, tokens, durations, and tool errors. Qwen 3.8 27B x-high on OpenCode achieved 95.64% in 8m48s, the best fast result, while GLM 5.3 Flash Max on OpenCode scored highest at 96.89% in 20m28s. Other runs included GLM 5.3 Flash, Luna 5.6, SOL 5.6, and Astra 6.0 across Codex Open, OMP Open, OpenCode, DSH, and PTC harnesses, with scores ranging from 78.54% to 96.89%.

Hacker News · AISep 8, 03:42 UTC · 8d agoAI research HN 104↑ · 61 comments

nvidia/Qwen3.8-Flash-Next-NVFP4 — new model trending #28 on Hugging Face

NVIDIA released an NVFP4 4-bit quantized build of Alibaba's Qwen3.8-Flash-Next, a 125B-parameter MoE vision-language model, via Model Optimizer.

The checkpoint quantizes Qwen3.8-Flash-Next — a hybrid-attention (Gated DeltaNet and Qwen Sparse Attention) Mixture-of-Experts model with 125B total and 6B activated parameters, plus 51B n-gram embeddings and 4B MTP — using NVIDIA Model Optimizer v0.46.0. NVFP4 benchmarks stay close to FP8: GPQA Diamond 91.5 vs 92.0, MMMU Pro 78.3 vs 77.1, Terminal-Bench 2.1 82.9 vs 83.3. It targets Blackwell B200/B300 GPUs, runs on vLLM, supports 262K context extendable to 1M tokens, and is licensed under the NVIDIA Open Model License with Qwen Community License 1.0.

Hugging Face trending modelsSep 2, 04:03 UTC · 14d agoModel release

DPRK APTs: Ted backdoor and curlRAT target South Korean media and automotive sectors

Rapid7 uncovered a DPRK-linked Linux toolkit using a HAProxy-embedded ted backdoor, SSH keylogger, and curlRAT against South Korean media and automotive firms.

Rapid7 Labs identified a previously undocumented framework attributed with medium confidence to DPRK actors, targeting South Korean automotive and media organizations likely since early 2025. The toolkit embeds a backdoor compiled into HAProxy 2.8.12 using its filter API, plus trojanized crond, agetty, atd, sshd, and polkitd, an SSH keylogger storing credentials under /var/lib/sshd/, and a curl-based RAT with a watchdog thread. It enables remote command execution, malicious script injection into served webpages (a watering-hole loop), credential harvesting, and long-term surveillance. Hardcoded C2s are associated with APT37 via ThreatFox, and exposed groupware portals and mail servers align with Kimsuky tradecraft; the initial access vector and any CVE remain unconfirmed.

Rapid7 BlogSep 4, 12:00 UTC · 12d agoThreat actor in the wild1

MicroHasTEE: Bare-Metal Haskell for Type-Level Peripheral Ownership on Armv8-M

MicroHasTEE expresses TrustZone secure and non-secure firmware as one typed Haskell program, catching peripheral-ownership inconsistencies at compile time.

MicroHasTEE is a multiparty Haskell framework that models both TrustZone firmware images as participants in a single typed program, using type-level capability ledgers to track peripheral acquisition, configuration, transfer, and finalization. MicroHs compiles the shared program twice to produce separate bare-metal Secure and Non-secure firmware images. The framework rejects inconsistent resource use, post-configuration attribution changes, wrong-domain callbacks, and calls to unregistered Secure services. A door-lock case study on an STM32U5 Nucleo board produced images of 232.7 KiB and 228.4 KiB of flash with roughly 220 KiB of SRAM per domain.

arXiv cs.CRSep 11, 08:34 UTC · 5d agoResearch1

StepAudio 3 Gen Technical Report

StepAudio 3 Gen unifies TTS, voice design, music, and sound effects via discrete autoregressive modeling over RVQ tokens.

StepAudio 3 Gen is a general-purpose audio generation model covering zero-shot TTS, voice design, vocal generation, sound effects, music, vibe speech, and mixed audio in one framework. It uses discrete autoregressive modeling over residual vector quantization (RVQ) tokens rather than the diffusion Transformer paradigm, with a StepAudio Tokenizer representing audio at 12.5 Hz in a shared 16x2048 residual code space. Key design principles include interference-aware progressive pretraining, an RVQ Adaptor for multi-codebook acoustic representations, and shared discrete autoregressive modeling. The model reports state-of-the-art performance on TTS and voice design while retaining strong generation across speech, vocals, sound effects, and music.

Hugging Face daily papersSep 10, 20:00 UTC · 5d agoAI research

Show HN: Nari Qwen3-TTS and Qwen3-ASR – High accuracy, low latency and cost

Nari Labs claims top Coval voice AI benchmark rankings with low-latency, low-cost Qwen3-ASR and Qwen3-TTS inference endpoints.

Nari Labs says its Qwen3-ASR Fast endpoint ranks #1 in Coval's time-to-final-segment latency (p50 44 ms) with 3.6% WER at $0.12/hour, behind only AssemblyAI Universal 3.5 Pro on accuracy. Its Qwen3-TTS Fast ranks #2 in time-to-first-audio (p50 63 ms) and #1 in WER at 3.8%, priced at $10 per 1M characters. The company reports beating the official Qwen3 TTS Flash Realtime endpoint (8.8% WER, 692 ms median TTFA) and Baseten's dedicated endpoint (6.0% WER, 101 ms). Public beta APIs are moving to paid general availability with $20 in credits for existing accounts.

Hacker News · securitySep 14, 16:07 UTC · 2d agoAI tools & infra HN 21↑ · 7 comments

StepAudio 3 Realtime Technical Report

StepAudio 3 Realtime debuts an audio-language model with Think-While-Speaking reasoning, delivering full-duplex voice dialogue with top benchmark results.

StepAudio 3 Realtime is an audio-language foundation model built around a continuous listen-converse-think-act loop for real-time spoken interaction. Think-While-Speaking runs private reasoning in parallel with speech, reaching a 73.0 macro average on StepAudioChat in reasoning mode. The model reports 90.6 on MMSU, 98.9 overall on the Artificial Analysis Full-Duplex Bench, and 56.0% macro task success on tau-Voice. An integrated Voice Agent handles asynchronous tool execution without disrupting dialogue flow.

Hugging Face daily papersSep 11, 20:00 UTC · 4d agoAI research

Everything in Moderation: Per-Domain Coverage Optima and Alignment-Resistant Domain Gaps in Multi-Domain Mid-Training

Controlled mid-training experiments on Qwen3-8B-Base find each domain has a 10-40% coverage optimum and domain gaps survive alignment SFT.

Using Qwen3-8B-Base (with a 4B replication) across five semantically rule-disjoint KOR-Bench domains, the authors train 30 data allocations spanning the five-domain simplex at five seeds each. All five domains show interior optima in the moderate 10-40% coverage band, and domain gaps persist after a fixed-budget compensatory SFT pass, which raises 116/120 cells yet bridges 0/240 pairs at a 5% threshold. Zero coverage collapses mid-training-only accuracy, though a FineWeb-Edu-only control shows the collapse is partly generic drift. The results argue mid-training data composition requires principled design rather than reliance on later alignment.

arXiv cs.AI / cs.LG / cs.CLSep 8, 17:31 UTC · 8d agoAI research1

Perplexity Details Its GPU Embedding Stack: How Ivy, Tulip and ROSE Serve pplx-embed

Perplexity details its GPU embedding serving stack (Ivy, Tulip, ROSE), which reuses LLM prefill/decode kernels, CUDA graphs, and LazyTensors to cut launch overhead.

Perplexity engineers published a deep dive on the serving infrastructure behind pplx-embed, used across Perplexity Search and its API platform. The stack comprises Ivy (Rust HTTP gateway), Tulip (gRPC scheduling and batching), and ROSE (Runtime-Optimized Serving Engine), which reuses LLM prefill and decode kernels rather than running a separate embedding engine. Optimizations include whole-model CUDA graphs with lazy capture and a LazyTensor abstraction that overlaps CPU batch preparation with in-flight GPU work. Benchmarks are reported against vLLM v0.22.0 in BF16, with FlashAttention 4 generally fastest but FlashInfer 3 winning on Qwen-based models at very long sequence lengths.

MarkTechPostSep 6, 03:20 UTC · 10d agoAI tools & infra1

Agnes-AI/Agnes-3.0-Flash — new model trending #30 on Hugging Face

Agnes AI releases open-weight Agnes-3.0-Flash Preview, a 33B multimodal model with 262k-token context under Apache 2.0.

Agnes AI released Agnes-3.0-Flash Preview, an open-weights multimodal checkpoint with 33B parameters and a 262,144-token context window under Apache 2.0. The model supports text, image, and video understanding, tool calling, and adjustable reasoning effort. The repo clarifies this preview checkpoint is distinct from the production/API Agnes 3.0 Flash model, which uses a different configuration with a 1M-token context window. Reported reference results include IFBench 74.20 and SciCode 38.08 against peers such as Qwen3.6-35B-A3B, Kimi K2.5, and MiniMax M3.

Hugging Face trending modelsSep 11, 15:50 UTC · 5d agoModel release

NVIDIA Details BioNeMo Inference Runtime (BioIR): 2.90x Higher Boltz-2 Folding Throughput and 58.5K Residues per GPU-Hour on 8xH100

NVIDIA released BioNeMo Inference Runtime (BioIR), an open-source PyTorch-compatible library delivering 2.90x higher Boltz-2 protein-folding throughput on 8xH100 GPUs.

NVIDIA detailed BioIR, a Python library that accelerates Boltz-2, OpenFold2, and OpenFold3 structure-prediction inference on NVIDIA GPUs while preserving standard PyTorch workflows. On a matched benchmark of 1,000 human dimers on 8xH100 80GB GPUs, BioIR delivered 58.5K folded residues per GPU-hour versus 20.2K for a torch.compile baseline, a 2.90x throughput gain. BioIR already powered the AlphaFold Database expansion, generating about 31 million candidate complexes across 4,777 proteomes, with 1.81 million released as high-confidence predictions. Extrapolated to 1 million targets, estimated folding energy drops from 35 MWh to 11 MWh at 8-GPU TDP equivalents.

MarkTechPostSep 10, 21:57 UTC · 5d agoAI tools & infra1

CoVeR: Coverage-Based Token Pruning for Multi-View 3D Reasoning in VLMs

CoVeR, a training-free coverage-based token pruner, preserves 93.5% of VLM 3D-reasoning performance using only about 8% of visual tokens.

Researchers introduce CoVeR, a deterministic, training-free selector that chooses visual tokens to cover every region of a multi-view 3D scene using only token coordinates. Unlike learned-importance and voxelization pruners, it enforces an exact per-scene token budget, avoids saturation plateaus, and prevents near-duplicate selections. Experiments across four vision-language models show it surpasses prior state of the art by 3.9 percentage points on average across three 3D reasoning benchmarks.

Hugging Face daily papersSep 7, 20:00 UTC · 8d agoAI research

NOAH: Learning the Full Patient Journey. A Longitudinal Multimodal Time-Aware Model for Representation and Forecasting

Researchers introduce NOAH, a time-aware generative transformer trained on 559 million MIMIC clinical events to forecast multimodal patient trajectories.

NOAH is a task-agnostic, time-aware generative transformer trained on over 559 million clinical events from 431,000 hospital visits by 299,000 patients across the MIMIC dataset family. It uses bidirectional time integration and a variational latent space to model the stochastic evolution of patient states, natively processing medical images, time-series signals, categorical events, and structured or unstructured clinical records. The model supports autoregressive forecasting with optional time control, zero-shot classification, and counterfactual intervention simulation, with strong probing performance across clinical outcomes, 15 ICD chapters, and 29 comorbidities.

Hugging Face daily papersSep 7, 20:00 UTC · 8d agoAI research1

ukisai/Swift-Qwen3.8-27b — new model trending #30 on Hugging Face

UkisAI releases Swift-Qwen3.8-27B, a Qwen3.8-27B derivative using 58.3% fewer thinking tokens with <1% performance loss and ~1.95x speed-up.

UkisAI released Swift-Qwen3.8-27B, a reasoning-efficient derivative of Qwen3.8-27B that cuts thinking-token usage by 58.3% while staying within 1% of base performance, yielding a 1.95x speed-up on several tasks. The model was fine-tuned by penalizing reasoning-marker tokens that trigger overthinking, plus a transfer component from BottleCap AI's ThinkingCap-Qwen3.6-27B. Benchmarks include GPQA-Diamond 88.28% (base 88.38%), MMLU-Pro 84.95% (base 85.47%), and AIME 2026 94.00% (base 98.67%), with mean-token reductions of roughly 27-46% across tests. GGUF weights are available on Hugging Face alongside enterprise licensing options.

Hugging Face trending modelsSep 8, 13:35 UTC · 8d agoModel release

Building AI to accelerate science and improve lives

Google highlights AI-for-science advances: AlphaGenome Atlas mapping 9 billion genetic variants, WeatherNext 3 weather model, and global health AI tools.

Google detailed AI advances across science and health, including AlphaGenome Atlas, which mapped all 9 billion possible single-letter genetic changes in the human genome and was made openly available. WeatherNext 3 delivers 50% more accurate precipitation forecasts a day or more ahead and is already in products. AlphaFold is used by 4 million researchers in 190 countries, TB chest X-ray screening has processed 25,000+ scans across six nations, and the diabetic retinopathy model has supported 1.15 million screenings. Google also released its AI & Economy ATLAS global usage insights.

Google · AISep 15, 16:00 UTC · 1d agoAI industry

NOAH: Learning the Full Patient Journey. A Longitudinal Multimodal Time-Aware Model for Representation and Forecasting

Researchers introduce NOAH, a generative time-aware transformer trained on 559 million MIMIC clinical events to model and forecast patient trajectories.

NOAH is a task-agnostic, time-aware generative transformer designed to represent and forecast the full multimodal patient journey across medical images, time-series signals, categorical events, and clinical text. It was trained on over 559 million clinical events from 431,000 hospital visits covering 299,000 patients in the MIMIC dataset family. The architecture combines bidirectional time integration with a variational latent space to capture continuous patient state evolution and clinical stochasticity. NOAH supports autoregressive forecasting with time control, zero-shot classification, and counterfactual intervention simulation, with evaluations on 15 ICD chapters, 29 comorbidities, and time-to-event prediction.

arXiv cs.AI / cs.LG / cs.CLSep 8, 17:56 UTC · 8d agoAI research2

New Ted Backdoor Hides Inside Victims' Own HAProxy Builds to Intercept Web Traffic

Rapid7 found a new backdoor, ted, compiled into trojanized HAProxy at two South Korean organizations, with medium-confidence attribution to North Korean actors.

Rapid7 documented a previously undocumented Linux toolkit named ted compiled into the HAProxy load balancer binaries of two South Korean organizations in the automotive and media sectors. The implant intercepts web traffic, serves altered pages only to filtered visitors, and hides C2 exchanges from backend logs and HAProxy statistics; a companion RAT, curlRAT, beacons on a default 12-hour schedule. The toolkit also trojanizes crond, sshd, agetty, atd, and polkitd binaries and sanitizes logs and bash history. Rapid7 attributes the activity with medium confidence to North Korean state-sponsored actors, with domain infrastructure overlapping APT37 listings in maltrail and delivery resembling the Operation SyncHole campaign.

The Hacker NewsSep 4, 14:51 UTC · 12d agoThreat actor in the wild

An Open-Source End-to-End FHE Implementation for Privacy-Preserving Llama 3 8B Inference

Odin runs Llama-3-8B fully homomorphic encrypted inference on a single H100 in 366 seconds, a 4.51x speedup over THOR.

Odin is an open-source end-to-end GPU CKKS implementation for privacy-preserving Llama-3-8B inference that co-designs ciphertext packing with model execution. A feature-major cross-layer layout unifies residual connections and layer interfaces, while transient intra-operator layouts serve linear projections and attention, avoiding intermediate repacking of QK^T softmax outputs. Minimax polynomial approximation with input-range control reduces polynomial degree and multiplicative depth for nonlinear ops. With 128-token input, Odin evaluates all 32 Transformer layers on one NVIDIA H100 80 GB in 366.4 s using 58.9 GiB peak memory, versus 1651.9 s for the THOR baseline, a 4.51x speedup.

arXiv cs.CRSep 11, 02:51 UTC · 5d agoResearch

NCP-ArchPreview Technical Report: Moving towards Latent Space Language Models through Next Concept Prediction

An 8.9B-parameter latent-space language model using next-concept prediction matches OLMo-3-7B pretraining loss with only 51.3% of the training tokens.

NCP-ArchPreview augments next-token prediction with Next Concept Prediction over a product-quantized concept vocabulary built from hidden states, trained jointly end-to-end. The 8.9B model was trained on 5.73T tokens from the Dolma-3 dataset, the largest latent-space language model demonstration to date. It consumes 51.3% of the tokens to reach OLMo-3-7B's final pretraining loss and outperforms it by 2.45 points on the downstream macro-average, including a 5.99-point GSM8K gain. The learned latent space also enables lightweight domain adaptation via a 17M-parameter VQ module and improves speculative drafting accepted length by 4.17%.

Hugging Face daily papersSep 8, 20:00 UTC · 7d agoAI research1

[AINews] DeepSeek v4.1-Flash: 763B-P8B-D16B novel causal Encoder–Decoder architecture with vision marks the Return of the Whale

DeepSeek released V4.1-Flash, an open-weight 763B-parameter model with a novel causal encoder-decoder architecture, 1M context, vision input, and MIT license.

DeepSeek launched V4.1-Flash, an open-weight MIT-licensed model using a novel causal encoder-decoder architecture with 763B total parameters and asymmetric active parameters: 8B for prefill and 16B for decode. It supports 1M-token context and text+image input, priced at $0.30 per 1M input and $1.20 per 1M output tokens with a 50% off-peak discount. Artificial Analysis scored it 40 on its Intelligence Index, above DeepSeek V4 Pro 0813, and Vals ranked it the #1 open-weight model ahead of Kimi K3. Baseten shipped day-0 support and Ollama began rolling it out to paid subscribers.

Latent SpaceSep 12, 05:56 UTC · 4d agoModel release 4 sources1

FlashVector: Agent for Hierarchical Model Serving Stack Optimization

FlashVector agent optimizes all layers of Unity's ad-serving stack, delivering up to 2x model-server throughput and 1.98x latency speedup in production.

FlashVector is an agentic system that optimizes performance across GPU kernels, ML framework computation graphs, model servers, and on-demand feature processing. Deployed in Unity's Vector advertising platform, it achieved up to 2x model-server throughput increase, 1.98x latency speedup, and 1.6x feature-store throughput gain. Optimizations spanned NVIDIA Triton's C++ codebase and the Python feature transformation service, demonstrating extensibility beyond single-kernel tuning.

arXiv cs.AI / cs.LG / cs.CLSep 15, 16:27 UTC · 1d agoAI research

GLM-5.3: How Chinese labs keep stride with the frontier

Z.ai released GLM-5.3, a ~750B-parameter model with frontier agentic coding scores, with open weights on Hugging Face planned in two weeks.

Z.ai announced GLM-5.3, initially available only in its coding plan, with API access and open Hugging Face weights promised within two weeks. The roughly 750B-parameter model, one-third the size of Moonshot AI's Kimi K3, surpasses Kimi K3 on many benchmarks and beats Claude Fable 5 or GPT-5.6-Sol on some, placing it at the frontier of agentic coding benchmarks. GLM-5.3 reuses the GLM-5.2 base model with substantially extended post-training based on more RL environments, more diverse tasks and more compute. The post also analyzes how Chinese labs keep pace with the frontier, arguing release speed matters more than distillation.

InterconnectsAug 14, 21:23 UTC · Aug 14, 2026Model release