ZeroHour
DashboardNewsCVEsExploitsIncidentsAIBriefAnalyticsEntitiesWatchlistIOCsAsk the feedSourcesFeeds & APIKiosk mode

Search: “Nex-N2.5-mini”

30 stories

Searchkeywordsmart
DomainAllCyberAIWindow24h3d7d30dAllIn the wildImportant onlyShow duplicates
CategoryAllModel release 2
SeverityAllcritical 0high 0medium 0low 0info 2

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

nex-agi/Nex-N2.5-mini — new model trending #30 on Hugging Face

Nex-AGI releases Nex-N2.5 agentic model family (mini, Pro, Max) with a 1.6-trillion-parameter MoE Max, open weights, and hosted access via OpenRouter.

Nex-AGI launched Nex-N2.5, a family of agentic models in mini, Pro, and Max sizes, with the Max version built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation and the company's first complete post-training effort at trillion-parameter scale. The models target long-horizon computer use, web browsing, and visually grounded agentic tasks, with expanded agent training environments. Reported benchmarks include Max scoring 86.1 on Terminal-Bench 2.1 and 65.7 on SWE-Bench Pro, trailing Claude Opus 5. Weights are being released openly on Hugging Face and ModelScope, with hosted access through OpenRouter.

Hugging Face trending models · 8d agoModel release1

nex-agi/Nex-N2.5-Pro — new model trending #30 on Hugging Face

Nex-AGI launches Nex-N2.5 agentic model family (mini/Pro/Max), with Max built on a 1.6-trillion-parameter MoE foundation.

Nex-AGI introduced Nex-N2.5, a next-generation family of agentic models in three sizes (mini, Pro, Max) focused on long-horizon agentic tasks including computer use, web browsing, and autonomous program execution. Nex-N2.5-Max is built on a 1.6-trillion-parameter text-only Mixture-of-Experts foundation, marking the company's first complete post-training effort at trillion-parameter scale. Weights will be released open-source on Hugging Face and ModelScope, with hosted access via OpenRouter. Benchmark comparisons against Claude Opus 5, GPT-5.6 Sol, Kimi-K3, GLM-5.3, DeepSeek-V4-Pro-0813, and Qwen3.8-Max show competitive scores on Terminal-Bench 2.1 and SWE-Bench Pro, though weights were listed as "coming soon" at publication.

Hugging Face trending models · 8d agoModel release1

OpenBMB Releases MiniCPM5-2B: A 2.52B Dense Model Averaging 53.9 Across 34 Benchmarks and Built to Run On Device

OpenBMB released MiniCPM5-2B, a 2.52B-parameter Apache 2.0 on-device model averaging 53.9 across 34 benchmarks, ahead of Qwen3.5-4B.

OpenBMB released MiniCPM5-2B, a 2,516,756,480-parameter dense LlamaForCausalLM model with grouped-query attention and a 131,072-token context, under Apache 2.0, runnable via vLLM, SGLang, llama.cpp, and Ollama. It averages 53.9 across 34 benchmarks versus 51.1 for Qwen3.5-4B, with strong tool-use (97.1 on tau2-Bench Telecom) and code results (69.1 LiveCodeBench v6, 46.4 SWE-bench Verified). Training combined 400B tokens of deep-thinking SFT, critic-based JustRL II RL teachers, and on-policy distillation merging 16 RL experts; datasets and intermediate checkpoints were published alongside the weights.

MarkTechPost · 9d agoModel release

openbmb/MiniCPM5-2B-GGUF — new model trending #30 on Hugging Face

OpenBMB released MiniCPM5-2B, a dense 2B on-device model claiming open-source SOTA among 2B-class models.

OpenBMB released MiniCPM5-2B, the second model in the MiniCPM5 series following MiniCPM5-1B, as a dense 2B Transformer built for on-device and resource-constrained deployment with GGUF weights on Hugging Face. The team claims 2B-class open-source state-of-the-art performance, remaining competitive with 4B-class models in coding, mathematics, long-context understanding, tool use and agentic tasks. The release includes a tech report, GitHub repository and online demo, and is currently trending on Hugging Face.

Hugging Face trending models · 12d agoModel release2

openbmb/MiniCPM5-2B — new model trending #4 on Hugging Face

OpenBMB released MiniCPM5-2B, a dense 2B open-weights Transformer claiming 2B-class open-source SOTA for on-device deployment.

OpenBMB released MiniCPM5-2B, the second model in the MiniCPM5 series following MiniCPM5-1B, and it is trending #4 on Hugging Face. The dense 2B Transformer targets on-device, local, and resource-constrained deployment and claims 2B-class open-source SOTA while remaining competitive with 4B-class models. Reported strengths include coding, mathematics, long-context understanding, tool use, and agentic tasks; a tech report (arXiv 2506.07900), GitHub repo, and online demo accompany the release.

Hugging Face trending models · 10d agoModel release

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 models · 14d agoModel release1

From Megawatts to Tokens: How NVIDIA Maximizes AI Factory Production

NVIDIA detailed DSX power-management results: Lambda gained 24% token throughput at fixed power, and an AI factory auto-shed 1MW via Emerald AI's grid program.

NVIDIA says Lambda's first validation of DSX MaxLPS on HGX B200 servers ran 19 nodes within a 16-node power budget, lifting cluster token throughput 24% (roughly 4M to 5M tokens/second) and improving performance per watt by 23%. NVIDIA projects DSX MaxLPS can enable up to 40% more GPU capacity for Vera Rubin NVL72 factories within the same megawatt budget. Emerald AI's Conductor platform, running at NVIDIA's Eos factory with Silicon Valley Power, responded to over 200 utility demand signals, automatically dropping power from 4MW to 3MW without interrupting priority workloads. The first dedicated DSX Flex commercial deployment is planned at a 96-megawatt Manassas, Virginia facility.

NVIDIA Blog · 1d agoAI industry

NVIDIA security advisory (AV26-849)

Canada's Cyber Centre relayed NVIDIA security advisories covering vulnerabilities in NemoClaw, OpenShell, Unified Fabric Manager, and DGX Spark.

The Canadian Centre for Cyber Security issued bulletin AV26-849 flagging NVIDIA vulnerabilities across NemoClaw and OpenShell (multiple versions), NVIDIA Unified Fabric Manager (multiple versions and models), and DGX Spark (versions prior to 1.110.13). The advisory links to NVIDIA security bulletins and urges users and administrators to apply updates as they become available. No exploitation details or CVE identifiers are enumerated in the advisory text.

Canadian Centre for Cyber Security · 21d agoAdvisory

Building a Linux GPU Driver for the M4 Mac Mini in One Month

Two developers built a fully OpenGL ES 3.0 compliant Linux GPU driver for the M4 Mac Mini in one month via clean-room reverse engineering.

Niklas and the author reverse engineered Apple's AGX GPU firmware ABI and user-space components in about a month, a process that normally takes years, producing an OpenGL ES 3.0 conformant driver fast enough to run Minecraft at 200fps on an M4 Mac Mini. The work was done transparently using hypervisor traces without examining Apple binaries, following clean-room practices, and included a custom shader compiler, command stream builder, and a full Linux kernel driver for the firmware ABI. The A18 Pro firmware ABI proved significantly more complex than the M1's, with 1.5x as many structs and twice as many pointers. All experiments and provenance evidence were published in public agx-re repositories.

Hacker News · AI · 1d agoResearchHN 40↑ · 5 comments

Arm Mali G2-Ultra NX GPU: desktop-class mobile gameplay with AI-native graphics

Arm unveiled Mali G2-Ultra NX, its first AI-native mobile GPU with in-shader neural acceleration, third-gen ray tracing, and up to 24% higher benchmark performance.

Arm announced the Mali G2-Ultra NX, the first AI-native Mali GPU, integrating neural accelerators directly into shader cores alongside a new execution engine and third-generation hardware ray tracing. It introduces Neural Super Sampling (NSS), Neural Frame Rate Upscaling (NFRU), and Neural Super Sampling and Denoising (NSSD); the Neural Dawn demo with Sumo Digital showed up to 4x performance efficiency and 70% lower external memory traffic versus native rendering. Arm claims up to 24% higher benchmark performance, 13% lower DRAM traffic on ray tracing benchmarks, and up to 120 FPS with NFRU. Over 14 billion Mali GPUs have shipped to date.

Hacker News · AI · 8d agoAI industryHN 71↑ · 53 comments1

With Groq 3 LPX in Full Production, NVIDIA Extends Vera Rubin Inference for Agents

NVIDIA puts Groq 3 LPX into full production and extends Vera Rubin NVL72 rack-scale systems for fast token generation in agentic AI inference.

NVIDIA announced that Groq 3 LPX is in full production as part of an extension of the Vera Rubin NVL72 rack-scale platform aimed at agentic AI inference. The announcement frames the next era of inference as full-stack AI factory co-design across chips, networking, and systems rather than a single component breakthrough. The focus is improving token generation speed for agent workloads.

NVIDIA Blog · 23d agoAI industry

AI Infra Summit: NVIDIA Vera Rubin and DSX Platform Advancements Showcase Energy Efficiencies of Optimizing Tokens Per Watt for AI Factories

At AI Infra Summit, NVIDIA showcased Vera Rubin and DSX gains up to 1.4x tokens per megawatt, plus Annapurna, d-Matrix, and Pinterest partnerships.

Ian Buck's AI Infra Summit keynote before 8,000+ attendees emphasized validated agentic tokens per megawatt as the emerging AI infrastructure metric. Announcements include Amazon's Annapurna Labs collaborating on NVHBM custom high-bandwidth memory, d-Matrix integrating NVLink Fusion with Raptor XPUs, and Pinterest using Blackwell plus Dynamo inference software for conversational visual discovery. Lambda reported 23% better performance per watt with DSX MaxLPS on Blackwell servers, running 19 nodes on a 16-node power budget. NVIDIA says DSX MaxLPS combined with Groq 3 LPX on Vera Rubin NVL72 targets up to 35X token throughput per megawatt versus GB200 NVL72 for 2-trillion-plus-parameter models.

NVIDIA Blog · 1d agoAI industry

d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment

d-Matrix will integrate its Raptor inference XPUs with NVIDIA NVLink Fusion, MGX racks and Spectrum-X networking for rack-scale AI factory deployment.

Inference chipmaker d-Matrix announced adoption of NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA's scale-up and scale-out networking, MGX rack architecture, and broader AI factory platform. NVIDIA claims 3x lower XPU-to-XPU latency than off-the-shelf Ethernet and 3 TB/s per-XPU all-to-all bandwidth via sixth-generation NVLink. d-Matrix plans to integrate Vera CPUs, ConnectX-9 SuperNICs, BlueField-4 DPUs and Spectrum-X Ethernet, with racks able to work alongside Vera Rubin NVL72 GPU systems. Other NVLink Fusion partners include AWS, Arm, Intel, Fujitsu, Marvell, MediaTek, Samsung and Cadence.

NVIDIA Blogupdated · 11h agofirst · 6d agoAI industry 2 sources

N4D Mesh Controller: New infrastructure, a UPX-packed agent labeled "go-titan," and how to hunt for it

Datadog researchers detonated a new N4D Mesh Controller sample in microVMs, revealing rotated infrastructure, a UPX-packed go-titan agent, and persistence behavior.

Datadog Security Research executed a newer N4D Mesh Controller sample in isolated microVMs, uncovering rotated command infrastructure and a UPX-packed agent labeled go-titan. Runtime evidence showed MCP tool abuse, multi-service scanning, and persistence mechanisms. The report provides hunting guidance for defenders tracking this infrastructure.

Datadog Security Labs · 28d agoMalware

N-able Releases Hotfix for Critical Remote Code Execution Vulnerability

N-able shipped Hotfix 4 patching CVE-2026-86218, a CVSS 10.0 pre-authentication RCE in N-central, with no confirmed production exploitation yet.

CVE-2026-86218 is a critical pre-authentication remote code execution flaw in N-able's N-central remote monitoring and management platform, disclosed September 6 with a maximum CVSS score of 10. It affects N-central versions before 2026.3.1.14 and is patched in N-central 2026.3 Hotfix 4. N-able says it has found no evidence of exploitation in production environments. It is the fifth N-able vulnerability disclosed in weeks, following two KEV-listed authentication bypasses and two internal API bypasses.

Infosecurity Magazine · 9d agoVulnerabilityCVE-2026-86218CVE-2026-18556CVE-2026-18577+2 CVEs

Training a 3.8B LLM to 0.384 CORE for $998 – Hugo Vergnes

Independent developer Hugo Vergnes trained a 3.8B-parameter Llama-style model to 0.384 CORE on 65B tokens for $998 in 43 hours on rented B200s.

Hugo Vergnes trained little-lm, a 3.848B-parameter decoder-only LLM, on 65.3B tokens in 43 hours for $998 using rented NVIDIA B200s, scoring 0.384 on the CORE benchmark and beating nanochat d32 (0.310) at similar cost. The Llama-style architecture uses RMSNorm, RoPE, GQA with 24 query and 8 KV heads, relu-squared MLPs, QK-norm, and ResFormer-style value embeddings that account for 19% of parameters. Key wins included the Muon optimizer for matrix parameters, a trapezoidal learning-rate schedule with linear cooldown, FP8 training plus vocabulary padding for roughly 33% throughput gains, and the ClimMix dataset over FineWeb-Edu. The project, inspired by Karpathy's nanochat, was built as a config-driven YAML framework for small LLM training.

Hacker News · security · 7d agoAI researchHN 28↑ · 2 comments1

N-able N-central HF4 Fixes Critical RCE After Series of Authentication Flaws

N-able released N-central 2026.3 Hotfix 4 (build 2026.3.1.14) fixing CVE-2026-86218, a critical pre-authentication RCE.

N-able shipped N-central 2026.3 Hotfix 4 (build 2026.3.1.14) to patch CVE-2026-86218, a critical remote code execution vulnerability exploitable without authentication. The hotfix follows a series of authentication flaws discovered in the N-central platform. Organizations running N-central should apply the update promptly; the notice does not report active exploitation.

SOCRadar · 9d agoVulnerabilityCVE-2026-86218

N-able patches critical N-central zero-day exploited in the wild (CVE-2026-86218)

N-able shipped an emergency hotfix for CVE-2026-86218, a pre-auth RCE zero-day in N-central RMM observed exploited in the wild.

N-able released Hotfix 4 for N-central 2026.3 on September 5 (build 2026.3.1.14), fixing CVE-2026-86218, a critical pre-authenticated remote code execution flaw in its RMM platform popular with MSPs. A private customer notice marked urgent said the zero-day 'has been observed being exploited in the wild,' while the public advisory said there were no confirmed production exploits. Huntress also flagged two high-severity vulnerabilities, CVE-2026-86206 and CVE-2026-86207, which allow authentication bypass and unrestricted access; they were patched the same weekend and were discovered by Rapid7's Stephen Fewer. N-able advised auditing N-central user accounts for unexpected users.

Help Net Security · 10d agoExploit / PoC in the wildCVE-2026-86218CVE-2026-86206CVE-2026-86207

Critical N-able N-central Vulnerability and Active Exploitation

N-able N-central pre-auth RCE zero-day CVE-2026-86218 (CVSS 10.0) is exploited in the wild; on-prem admins must upgrade to 2026.3 HF4.

N-able disclosed a third N-central vulnerability, CVE-2026-86218, a pre-authentication RCE rated CVSS 10.0, and released hotfix 2026.3 HF4 superseding build 2026.3.1.13. Huntress reproduced an exploit chain involving an authentication bypass (CVE-2026-86206/CVE-2026-86207) after a fully patched customer's N-central production server was compromised on September 4. Attackers appended strings like .invalid to account names and probed the /remoteControlAction.do?method=getPierDetails endpoint; Huntress worked with Cloudflare to disable adversary tunnel infrastructure. Hosted N-central instances are already patched; on-prem administrators must upgrade immediately.

Huntress · 10d agoExploit / PoC in the wildCVE-2026-86218CVE-2026-86206CVE-2026-86207+2 CVEs

H Company Releases NeoMME: A Family of 260M and 800M Single-Tower Multimodal Encoders That Drop the Vision Tower and Causal Decoder

H Company released NeoMME, 260M/800M single-tower multimodal encoders matching 3.75B ColQwen2.5 on ViDoRe v3 while being 14.4x smaller, under Apache 2.0.

H Company released NeoMME, a family of 262,937,906- and 793,715,032-parameter bidirectional encoders that process text and raw 32x32 image patches in a single tower, pretrained via masked diffusion and released under Apache 2.0 with day-zero Hugging Face Transformers support. NeoMME-Retriever-260M reaches 0.523 nDCG@10 on ViDoRe v3, matching 3.75B-parameter ColQwen2.5 while being 14.4x smaller; the 800M model scores 0.556. Hierarchical token pooling with int8 and binary quantization shrinks late-interaction indexes from roughly 1.5 MB to 6 kB per page while retaining 95.19% of nDCG@10; text-only BEIR retrieval remains a weak spot.

MarkTechPost · 10d agoAI research

N-able Patches Critical Zero-Day in N-central

N-able patches critical unauthenticated RCE zero-day CVE-2026-86218 in N-central, exploited in the wild; on-premises admins must apply hotfix 2026.3 HF4.

N-able released an urgent hotfix (2026.3 HF4) for CVE-2026-86218 (CVSS 10), an unauthenticated RCE in N-central exploited as a zero-day. Scanning/exploitation attempts observed from IP range 23.234.64.0/18 starting September 4; admins should check logs for scanning and unrecognized new accounts. The hotfix supersedes patches for CVE-2026-86206 and CVE-2026-86207, which Huntress observed potentially chained in the wild to bypass authentication in production environments.

SecurityWeek · 8d agoExploit / PoC in the wildCVE-2026-86206CVE-2026-86207CVE-2026-86218

[AINews] NVIDIA buys HuggingFace for $13B, as OpenAI publishes their HF incident retro

Z.ai released open-weight GLM-5.3-Flash (320B/18B active, 1M context, MIT) while Nvidia confirmed buying Hugging Face for $13B.

Z.ai formally launched GLM-5.3-Flash, the model previously previewed as Ox Alpha: 320B total parameters with 18B active, a 1M-token context window, natively multimodal, MIT-licensed, and claimed on par with Claude Opus 4.8 on coding. Artificial Analysis scored it 57 on its Intelligence Index at $0.09 per task, roughly 7.5x cheaper than GLM-5.3, and it scored 84.3% on Terminal-Bench 2.1. Nvidia's $13B acquisition of Hugging Face (~80x its $150M ARR) was confirmed, nearly double its initial $7B January offer. The roundup also notes Qwen shipping an impressive Flash model on Chinese chips as part of a broader open-model narrative.

Latent Space · 21d agoModel release1

N-able ships second N-central hotfix as attackers keep exploiting CVE-2026-18577

N-able shipped a second N-central hotfix as attackers actively exploit authentication bypass CVE-2026-18577, attributed to ransomware group Storm-1175.

N-able released Hotfix 2 (N-central 2026.3.1.10) to counter ongoing exploitation of CVE-2026-18577, an authentication bypass that evades the patch for the earlier CVE-2026-18556; exploitation was first detected on August 1, 2026. Post-exploitation includes using the Take Control feature to reach managed endpoints, registering a Cloudflare tunnel service for persistence, creating a 'veeam' domain account, resetting admin passwords, and disabling Microsoft and Sophos security tooling. Microsoft analysts link the activity to Storm-1175, which now deploys a new StormEncryptor ransomware strain instead of Medusa, often reaching data exfiltration and ransomware within days. Sophos and Huntress expanded the IOCs and warned that partners who patched late should treat environments as potentially compromised.

Help Net Security · Aug 12, 2026Exploit / PoC in the wildCVE-2026-18577CVE-2026-185561

Up to 30x More Work Per Watt: NVIDIA Vera Rubin NVL72 Sets a New Efficiency Standard for AI Agents

NVIDIA claims Vera Rubin NVL72 delivers up to 30x more work per watt, citing OpenRouter data that agentic workloads use 15x more tokens than chat.

NVIDIA positions the Vera Rubin NVL72 as a new efficiency standard for AI agents, claiming up to 30x more work per watt. The company cites OpenRouter data showing agentic AI workloads consume 15x more tokens than a simple chat request, using a financial-research agent example that spawns sub-agents and multiple tool calls. The piece is largely a product efficiency narrative rather than independent benchmarking.

NVIDIA Blog · 23d agoAI industry

NVIDIA and CrowdStrike Strengthen Agentic Cybersecurity Frontier

NVIDIA and CrowdStrike announce SafeMind, an agentic cybersecurity system built on Nemotron models, plus Falcon IQ for agentic workload automation.

At CrowdStrike's Fal.Con 2026, NVIDIA and CrowdStrike announced SafeMind, an agentic cybersecurity system combining CrowdStrike's purpose-built frontier models and harnesses with NVIDIA Nemotron open models in a continuous red-versus-blue coevolution loop. A Blue Solano model post-trained on Nemotron 3 Super reportedly achieved higher accuracy than leading frontier models at 99% lower cost. CrowdStrike also introduced Falcon IQ, powered by Nemotron models in the Charlotte AI AgentWorks platform, coordinating more than 50 agents for automated defensive workflows. CrowdStrike cited an 89% year-over-year rise in AI-enabled attacks and a fastest eCrime breakout time of 27 seconds as context for agentic defense.

NVIDIA Blog · 15d agoAI industry

Nvidia and Palantir team up to run supply chains with AI, starting with Nvidia's own million-part operation

Nvidia and Palantir integrated open Nemotron models and cuOpt into Foundry to run AI-driven supply chains, starting with Nvidia's million-part operation.

Nvidia and Palantir announced a partnership to run supply chains with AI, first deployed on Nvidia's own network of millions of parts and thousands of suppliers; a single Vera Rubin rack contains 1.3 million components. Palantir is integrating open Nemotron models into its Foundry platform for customer fine-tuning, while Nvidia cuOpt handles scenario planning and optimization, with decisions fed back to improve models over time. The stack runs on customer hardware or in the cloud through the Sovereign AI OS reference architecture with infrastructure partners Dell, Cisco, Rackspace, and Nebius, with more details shown at AIPCon 11.

The Decoder · 6d agoAI industry1

Back-to-back N-able bugs send admins on a patching spree

CVE-2026-86218, a CVSS 10.0 pre-auth RCE in N-able N-central, is being exploited in the wild; Hotfix 4 mitigates it immediately.

N-able disclosed CVE-2026-86218 on September 6, a pre-authentication remote code execution flaw with CVSS 10.0 in its N-central RMM platform, and both N-able and Huntress report it is being exploited in the wild. It follows Huntress's disclosure of an exploit chain combining CVE-2026-86206 and CVE-2026-86207 that bypasses access controls to create unauthorized administrative accounts, investigated after a September 4 compromise of a fully patched customer environment. N-able has applied mitigations to all hosted N-central instances; on-premises customers must upgrade to Hotfix 4 (build 2026.3.1.14) immediately.

CSO Online · 9d agoExploit / PoC in the wildCVE-2026-86218CVE-2026-86206CVE-2026-86207+2 CVEs

Cohere Releases North Small Translate: A 218B MoE Translation Model That Scores 83.6 on WMT26 Across 50 Languages

Cohere released North Small Translate, an open-weight 218B MoE (25B active) translation model scoring 83.6 on WMT26 across 50 languages.

Cohere and Cohere Labs released North Small Translate, a decoder-only sparse Mixture-of-Experts translation model with 218B total and 25B active parameters, 128 experts with 8 activated per token plus shared experts, and 16K-token input and output context. In Cohere's vendor-reported WMT26 evaluation, judged by GPT-5.6-Sol, it scores 83.6 averaged across 50 languages (84.36 in an agentic multi-pass mode), ahead of DeepL NextGen (81.37), Qwen 3.5 397B A17B (81.56), GLM 5.2 (76.50), and Google Translate (68.20). The model was built with RWS's Language Weaver team, post-trained specifically for translation, and reports 112 output tokens per second versus 81 for Gemma 4 31B, with long-document xCOMET-XL scores of 48.9 versus 21.3 for Google Translate. It is available free on Cohere's Chat V2 API until rate limits, with three self-hosting checkpoints including a 4-bit NVFP4 variant running on 1x B200 or 2x H100.

MarkTechPost · 5d agoModel release

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 Blog · 13d agoAI industry

Iris-mini and Iris-pro are the strongest open-weight search agents in their class

Chinese lab AllSpark releases Iris-mini (35B) and Iris-pro (397B) open-weight search agents claiming best-in-class results on BrowseComp and other research benchmarks.

AllSpark's paper introduces Iris-mini (35B parameters, built on Qwen3.6-35B-A3B) and Iris-pro (397B parameters, built on Qwen3.5-397B-A17B), both with 256,000-token context windows. Iris-pro scores 88.6 on BrowseComp, 85.1 on BrowseComp-ZH, 92.9 on DeepSearchQA, and 56.4 on Humanity's Last Exam; Iris-mini reaches 82.2, 84.8, 86.9, and 52.3 respectively. Training tasks are reverse-engineered from web link structure, filtered by a judge model, and refined via alternating SFT and reinforcement learning ('SFT-RL climbing') against live web search. Weights are available on Hugging Face, and the Iris Harness with agent loop, tools, and all four benchmarks is on GitHub.

The Decoder · 3d agoModel release1