AGI Pulse

Latest update history

Browse the latest-update history by date. 46 in this window; filter by date or type.

Digests

2026-10-225 items

10-22
Canva launches Craft, its first in-house image model: DesignArena #4 debut, two-stage plan-then-render, live in Dream Lab and Public API
Read original →
Model releaseCanvaagipulse
10-22
Hugging Face open-sources Carbon-X, a 14B Apache-2.0 model you can chat with about DNA: averages 66.7% on five DNA benchmarks vs GPT-6.1 Sol's 67.5% at ~1/1,800 the cost, runs on one 80 GB GPU
Read original →
Model releaseHugging Faceagipulse
10-22
miHoYo and Tsinghua open GameScaling: eight Apache-2.0 game-playing models (0.8B–27B) that output keyboard-and-mouse actions; 27B tops Minecraft specialists and 9B beats frontier models in real-time play
Read original →
Model releasemiHoYoagipulse
10-22
iFlytek launches Spark X2.6: a 293B-A30B MoE with 1M context that it says was trained entirely on Chinese chips, priced at ¥0.9/¥3.6 per million tokens; only the 4B sibling has open weights
Read original →
Model releaseiFlytekagipulse
10-22
Baseten's Base Labs open-sources decision models mammoth-1 (27B) and elephant-1 (9B) under MIT: mammoth-1 tops the Jev Decision Index at 71.24, elephant-1 ties Liquid's closed d1-ultra on Artificial Analysis
Read original →
Model releaseBasetenagipulse

2026-10-212 items

10-21
NVIDIA opens Kumo foundation models for tabular business data and RFD4-Proteina, a protein design model with the Baker Lab it says is 13x faster than RFdiffusion3
Read original →
Model releaseNVIDIAagipulse
10-21
PUBG maker KRAFTON open-weights Raon-1 mini preview, its first LLM trained from scratch: a 30B-A4B MoE built on Kimi- and DeepSeek-derived attention, Apache 2.0
Read original →
Model releaseKRAFTON AIagipulse

2026-10-2013 items

10-20
humans& releases Ampersand v0.1, a 'collaborator' model RL-trained from GLM 5.3 to help groups reach shared goals, via an Agents API
Read original →
Model releasehumans&agipulse
10-20
ElevenLabs launches Eleven Realtime: v1 Mini and v1 Max speech-to-speech models for voice agents, with an OpenAI-compatible Realtime API
Read original →
Model releaseElevenLabsagipulse
10-20
EAGLE-4 speculative decoding goes public: one draft model that adapts to batch size and hardware, up to 3.6x faster on Qwen3.8-27B in the team's tests
Read original →
PaperEagle AI Corpagipulse
10-20
UCSD's FastVideo open-sources FastH3 Omni, a distilled MiniMax-H3 that makes reference-guided video with sound up to 12x faster and runs on an RTX 5090
Read original →
Model releaseHao AI Lab (UC San Diego)agipulse
10-20
NVIDIA releases Cosmos3-H-Dreams, an open real-time surgical world model trained on clinical gallbladder surgery that runs at 30 fps on one GPU
Read original →
Model releaseNVIDIAagipulse
10-20
NVIDIA GEAR releases Flora, open models that caption and score robot and human videos to train GR00T; weights are non-commercial
Read original →
Model releaseNVIDIAagipulse
10-20
Gradium open-sources Phonon, on-device text-to-speech models under 50M parameters that run with no GPU; only the English 31m model is licensed for commercial use
Read original →
Model releaseGradiumagipulse
10-20
Bria releases Fibo 2.0, an 8B image model that Artificial Analysis ranks #1 among open-weights image editors; weights are non-commercial only
Read original →
Model releaseBria AIagipulse
10-20
Mistral launches Voxtral 3: 50-language transcription, expressive speech in 35+ languages and a LiveKit-based voice agent SDK, API and self-hosted only
Read original →
Model releaseMistral AIagipulse
10-20
MemTensor open-sources MemT-1.0, memory-native models built on DeepSeek-V4-Flash and Qwen3.6 that keep a fixed-size memory inside the model instead of resending history; text-memory systems still lead overall
Read original →
Model releaseMemTensoragipulse
10-20
Microsoft Research releases a public preview of BioEmu2, an MIT-licensed open-weights foundation model for biomolecular structure and dynamics covering proteins, nucleic acids and ligands
Read original →
Model releaseMicrosoft Researchagipulse
10-20
Skyfall AI unveils SKY-001, an "engineering world model" for CAD and BIM design: it claims 51-125x the speed of GPT-6 Astra and other frontier models at 150-210x lower cost, early access by waitlist only
Read original →
Model releaseSkyfall AIagipulse
10-20
Ant Group's Ant Ling open-sources Ming-3.0, five MIT-licensed speech models: a 0.5B TTS that tops Artificial Analysis' open-weights voice-cloning arena, a 126B-A6.5B audio-understanding MoE and a dialect-aware ASR
Read original →
Model releaseAnt Group (inclusionAI)agipulse

2026-10-1915 items

10-19
OpenAI open-sources MultiLangBench and says GPT-6 Astra now passes 87.4% of SQLite-from-scratch tests across 20 languages, with GPT-6.1 Sol reaching 97.9% on COBOL migration
Read original →
PaperOpenAIagipulse
10-19
Viggle releases JST-2, a physics-native world model that builds 3D scenes in about a second and rig-free animated characters, free in its PINOC app and via API
Read original →
Model releaseViggleagipulse
10-19
Amnon Shashua's doubleAI launches Maxwell, a 'code execution model' that rewrites running CPU and GPU code for speed and formally proves each change
Read original →
Model releasedoubleAIagipulse
10-19
Ex-NVIDIA GR00T team's Dream Labs launches Dream-1 robot model with a public API from $3/hour and a $90M seed round
Read original →
Model releaseDream Labsagipulse
10-19
Meta publishes the Muse Glimmer 30B technical report: a dense model distilled from Muse Spark for a 32 GB budget, leading similar-size open models on tool use
Read original →
PaperMetaagipulse
10-19
Black Forest Labs and fal launch FLUX 3 Max video model: 10-second clips in about 11-12 seconds, joint #1 in Artificial Analysis' preliminary image-to-video ranking
Read original →
Model releaseBlack Forest Labs / falagipulse
10-19
Cloudflare previews Clairic, a System One extraction model on Workers AI that returns typed values, estimates and forecasts in one pass without generating tokens; open weights planned
Read original →
Model releaseCloudflareagipulse
10-19
Inception releases Mercury 3, a diffusion LLM it says beats GPT-6 Luna and Haiku 5.5 at 1,000+ tokens/s with 1M context; $0.20/$0.70 per 1M tokens, half off at launch
Read original →
Model releaseInceptionagipulse
10-19
MiniMax officially launches M3.1 Flash, reveals Space Bunny as its early version, publishes $0.15/$0.60 pay-go pricing and agentic benchmarks; open weights promised later
Read original →
Model releaseMiniMaxagipulse
10-19
GoTo releases a 27B Indonesian-language model with open weights on Hugging Face: Gemma 3 continued pre-training with AI Singapore, six languages incl. Javanese, Sundanese, Batak and Balinese, 128K context
Read original →
Model releaseGoToagipulse
10-19
ACE Robotics releases ACE-Brain-0.6, a 9B embodied model fine-tuned from Qwen3.5-9B on ~1M grounding samples, claiming to beat GPT-6 Astra on six of eight spatial-grounding benchmarks
Read original →
Model releaseACE Roboticsagipulse
10-19
MiniMax makes M3.1 Flash generally available in its API (1M context, multimodal, five thinking-effort levels) and starts retiring M2.5/M2.7, routing plan users to the new model
Read original →
Model releaseMiniMaxagipulse
10-19
VAST open-sources EdenLive: real-time generative rendering where a game engine keeps world state and a 5B/14B video model paints the frames, 24+ FPS at 720P on one RTX 5090 (self-reported)
Read original →
Model releaseVASTagipulse
10-19
Alibaba Tongyi releases OpenGCUA, an open computer-use agent stack: code-first harness lifts all eight frontier models on OSWorld-v2 (mean 45.3→64.9, 53% fewer steps), with trajectories released
Read original →
PaperAlibaba Tongyiagipulse
10-19
Tencent Hunyuan open-sources Hy Memory 2.0: a ~3B-active memory-writer model (Hy-Memory-30B-A3B) that edges Claude Opus 5 in its own memory benchmarks, plus an SDK and MemHub cross-tool memory app
Read original →
Model releaseTencent Hunyuanagipulse

2026-10-182 items

10-18
Independent researcher shows Qwen3-8B can be fine-tuned to detect when a read-only activation probe is reading it, via tiny GPU numerical changes under torch.compile; code, data and weights released
Read original →
PaperSauers (independent)agipulse
10-18
Google researchers open-source Harness², which lets an agent rewrite its own prompts, tools and memory without touching model weights: one step lifts GLM-5 and Gemini 3.5 Flash by up to 11.5 points on WorkBuddy-Bench
Read original →
PaperGoogleagipulse

2026-10-173 items

10-17
NVIDIA releases Nemotron Labs Dory1.0-4B on Hugging Face, a looped 4B model whose single checkpoint trades inference loops for accuracy (SWE-bench Verified 50.3→61.8 at 1→4 loops, self-reported; 256K context, OpenMDW-1.1)
Read original →
Model releaseNVIDIAagipulse
10-17
OpenMOSS open-sources MOSS-Image, a 5B DiT text-to-image model (Base + SFT, Apache 2.0) built for Chinese/English text rendering and JSON layout control
Read original →
Model releaseOpenMOSSagipulse
10-17
DeepSeek upgrades V4.1 Flash to V4.1-Flash-1017 with open MIT weights: Terminal-Bench 4.0 jumps from 31.2 to 42.3, same API price
Read original →
Model releaseDeepSeekagipulse

2026-10-166 items

10-16
Microsoft Research in Nature: Probabilistic Bias Correction doubles ECMWF AI model skill for 2–6 week forecasts, improves 91–98% of operational targets, and won every variable in ECMWF's 2025 real-time contest; open source
Read original →
PaperMicrosoft Researchagipulse
10-16
Perplexity open-sources pplx-decider v1.2 decision models under Apache 2.0: a 27B (Decision Index 68.1, up from 62.3) and a 4B Flash, all questions in one pass for up to 5x faster inference, at 2 and 1 cents per million tokens
Read original →
Model releasePerplexityagipulse
10-16
Google Research: AI agents run closed-loop wet-lab experiments, with a lab digital twin that blocks unsafe robot commands, aptamer hit rates up to 42% after three rounds, and nanomolar protein binders
Read original →
PaperGoogle Researchagipulse
10-16
Elorian, founded by ex-DeepMind researcher Andrew Dai, launches Glimpse, a visual-thinking model that zooms and rewatches video while it reasons; self-reported wins over GPT-6 Astra on some benchmarks at $0.60/$3 per million tokens
Read original →
Model releaseElorianagipulse
10-16
Liquid AI launches d1-ultra, an API-only multimodal decision model: #1 on the Artificial Analysis Decision Index (65), ahead of Jev 1.14, at $0.042 per million input tokens
Read original →
Model releaseLiquid AIagipulse
10-16
Google DeepMind: four Gemini 4 Argon agents beat Minecraft from a random spawn in 27 hours, writing 2,783 programs themselves with only standard Mineflayer; a single run on Easy
Read original →
PaperGoogle DeepMindagipulse

Load previous week →