Goblin
News
AI news, distilled.
|
News
About
News
About
Titles
Summaries
7
xAI Launches Grok Voice Transcribe 2.0 — Ranks #1 for Accuracy Among 32 Streaming STT Models
Models
2
23h ago
7
xAI Launches Grok Voice Transcribe 2.0 — Ranks #1 for Accuracy Among 32 Streaming STT Models
Models
· 2 srcs · 23h ago
Discuss
7
PrismML Releases Bonsai 2 27B: Ternary-Compressed Reasoning Model That Runs on Consumer Hardware
Updated Sep 18
Models
3
15h ago
7
PrismML Releases Bonsai 2 27B: Ternary-Compressed Reasoning Model That Runs on Consumer Hardware
Upd Sep 18
Models
· 3 srcs · 15h ago
Discuss
7
Alibaba Releases Qwen3.8-Omni-Flash: Native Omnimodal Model with 1M Context Window, Outpacing Gemini 3.8 Flash on Audio
Models
1
1d ago
7
Alibaba Releases Qwen3.8-Omni-Flash: Native Omnimodal Model with 1M Context Window, Outpacing Gemini 3.8 Flash on Audio
Models
· 1 src · 1d ago
Discuss
7
Stealth AI Model 'Union Alpha' Appears on OpenRouter with Competitive Coding Benchmarks
Models
2
1d ago
7
Stealth AI Model 'Union Alpha' Appears on OpenRouter with Competitive Coding Benchmarks
Models
· 2 srcs · 1d ago
Discuss
7
Kimi K3, the World's Largest Open-Weight Model, Launches on Amazon Bedrock
Updated Sep 19
Models
2
15h ago
7
Kimi K3, the World's Largest Open-Weight Model, Launches on Amazon Bedrock
Upd Sep 19
Models
· 2 srcs · 15h ago
Discuss
7
Figure's Helix 2.5 Humanoid Robot Deployed Zero-Shot Across 30 Bay Area Homes
Models
1
1d ago
7
Figure's Helix 2.5 Humanoid Robot Deployed Zero-Shot Across 30 Bay Area Homes
Models
· 1 src · 1d ago
Discuss
7
DeepSeek Releases V4.1 Flash with 4x KV Cache Compression for Long-Context Agent Workflows
Models
1
2d ago
7
DeepSeek Releases V4.1 Flash with 4x KV Cache Compression for Long-Context Agent Workflows
Models
· 1 src · 2d ago
Discuss
6
IFM Releases K2-Horizon-7B-Uno: Diffusion-Augmented LLM Claims 5,255 TPS Without Quality Loss
Models
1
2d ago
6
IFM Releases K2-Horizon-7B-Uno: Diffusion-Augmented LLM Claims 5,255 TPS Without Quality Loss
Models
· 1 src · 2d ago
Discuss
7
Periodic Neon LLM Outperforms GPT-6 Astra and Claude Fable 5.1 on Scientific XRD Analysis
Models
1
3d ago
7
Periodic Neon LLM Outperforms GPT-6 Astra and Claude Fable 5.1 on Scientific XRD Analysis
Models
· 1 src · 3d ago
Discuss
7
Odyssey Launches Odyssey-3, a Foundation World Model Controlling Robots, Vehicles, and Drones from One Architecture
Models
1
3d ago
7
Odyssey Launches Odyssey-3, a Foundation World Model Controlling Robots, Vehicles, and Drones from One Architecture
Models
· 1 src · 3d ago
Discuss
8
TypeSafe AI Launches Jev, a Frontier Decision Model Claiming 200x Speed and Free Output Tokens
Updated Sep 18
Models
7
1d ago
8
TypeSafe AI Launches Jev, a Frontier Decision Model Claiming 200x Speed and Free Output Tokens
Upd Sep 18
Models
· 7 srcs · 1d ago
Discuss
8
Google DeepMind Launches Gemini 3.8 Live Voice Models, Claims #1 Speech-to-Speech Benchmark
Updated Sep 19
Models
7
23h ago
8
Google DeepMind Launches Gemini 3.8 Live Voice Models, Claims #1 Speech-to-Speech Benchmark
Top
Upd Sep 19
Models
· 7 srcs · 23h ago
Discuss
8
Salesforce Launches Koa: First Open-Weight Reasoning Model Built on Nvidia Nemotron for Enterprise CRM
Updated Sep 17
Models
5
2d ago
8
Salesforce Launches Koa: First Open-Weight Reasoning Model Built on Nvidia Nemotron for Enterprise CRM
Top
Upd Sep 17
Models
· 5 srcs · 2d ago
Discuss
6
Nari Labs Launches Qwen3-TTS and Qwen3-ASR, Leading Coval's Voice AI Benchmarks
Models
1
5d ago
6
Nari Labs Launches Qwen3-TTS and Qwen3-ASR, Leading Coval's Voice AI Benchmarks
Models
· 1 src · 5d ago
Discuss
8
Anthropic, Google, and OpenAI Each Ship Dual-Tier Flagship Models: Public Paid and Identity-Gated Restricted Versions
Models
1
5d ago
8
Anthropic, Google, and OpenAI Each Ship Dual-Tier Flagship Models: Public Paid and Identity-Gated Restricted Versions
Top
Models
· 1 src · 5d ago
Discuss
7
Shanghai AI Lab Quietly Releases Atria Dawn Preview: 744B MoE Agentic Model with 256K Context
Models
1
5d ago
7
Shanghai AI Lab Quietly Releases Atria Dawn Preview: 744B MoE Agentic Model with 256K Context
Models
· 1 src · 5d ago
Discuss
6
Sakana AI Releases Fugu Ultra v2: Learned Multi-Agent Orchestration With 1M Token Context
Models
1
5d ago
6
Sakana AI Releases Fugu Ultra v2: Learned Multi-Agent Orchestration With 1M Token Context
Models
· 1 src · 5d ago
Discuss
7
InternLM Releases Intern-S2-397B Multimodal Model for Scientific Intelligence and Long-Horizon Agents
Models
3
6d ago
7
InternLM Releases Intern-S2-397B Multimodal Model for Scientific Intelligence and Long-Horizon Agents
Models
· 3 srcs · 6d ago
Discuss
7
Cohere Releases North Small Translate: MoE Model Tops DeepL and Google Translate on WMT26
Models
2
Sep 11
7
Cohere Releases North Small Translate: MoE Model Tops DeepL and Google Translate on WMT26
Models
· 2 srcs · Sep 11
Discuss
8.05
OpenAI Releases GPT-Image-2.5 Flare and Sunburst with Context-Aware Editing and 50% Lower Latency
Models
1
Sep 10
8.05
OpenAI Releases GPT-Image-2.5 Flare and Sunburst with Context-Aware Editing and 50% Lower Latency
Models
· 1 src · Sep 10
Discuss
8.05
Cognition Launches SWE-2: Advanced Coding AI Matches Frontier Models at a Fraction of the Cost
Updated Sep 12
Models
3
Sep 12
8.05
Cognition Launches SWE-2: Advanced Coding AI Matches Frontier Models at a Fraction of the Cost
Upd Sep 12
Models
· 3 srcs · Sep 12
Discuss
8
OpenAI Launches GPT-Live-1 in API: Full-Duplex Voice Model with Telephony Support at $0.05/min
Updated Sep 12
Models
4
Sep 12
8
OpenAI Launches GPT-Live-1 in API: Full-Duplex Voice Model with Telephony Support at $0.05/min
Top
Upd Sep 12
Models
· 4 srcs · Sep 12
Discuss
6
GigaChat 3.5 Reasoning 438B-Parameter Model Released on Hugging Face
Models
1
Sep 10
6
GigaChat 3.5 Reasoning 438B-Parameter Model Released on Hugging Face
Models
· 1 src · Sep 10
Discuss
6
IBM Releases Granite Time Series PatchTST-FM-r2, Top Commercial-Licensed Zero-Shot Forecasting Model
Models
1
Sep 9
6
IBM Releases Granite Time Series PatchTST-FM-r2, Top Commercial-Licensed Zero-Shot Forecasting Model
Models
· 1 src · Sep 9
Discuss
7
Inception Releases Mercury 2.5 Diffusion LLM with 40% Intelligence Boost and 1,107 Tokens/Second
Models
1
Sep 8
7
Inception Releases Mercury 2.5 Diffusion LLM with 40% Intelligence Boost and 1,107 Tokens/Second
Top
Models
· 1 src · Sep 8
Discuss
7
inclusionAI Releases Ling-3.0-Flash-VL: 124B Multimodal Model with 1M-Token Context and Video Reasoning
Models
1
Sep 8
7
inclusionAI Releases Ling-3.0-Flash-VL: 124B Multimodal Model with 1M-Token Context and Video Reasoning
Models
· 1 src · Sep 8
Discuss
6
Qwen Team Releases Qwen-Drive-1.0: Open Autonomous Driving VLM with 3D Perception and Motion Planning
Models
2
Sep 8
6
Qwen Team Releases Qwen-Drive-1.0: Open Autonomous Driving VLM with 3D Perception and Motion Planning
Models
· 2 srcs · Sep 8
Discuss
6
OUI-1: First Open-Weight Model for Generative UI Released
Models
1
Sep 8
6
OUI-1: First Open-Weight Model for Generative UI Released
Models
· 1 src · Sep 8
Discuss
7
OpenAI President Greg Brockman Discusses Astra Launch, Alignment, and AI Value Chain
Models
1
Sep 7
7
OpenAI President Greg Brockman Discusses Astra Launch, Alignment, and AI Value Chain
Models
· 1 src · Sep 7
Discuss
7
GPT-6 Astra Mines a Minecraft Diamond Overnight in Autonomous Computer-Use Demo
Models
1
Sep 7
7
GPT-6 Astra Mines a Minecraft Diamond Overnight in Autonomous Computer-Use Demo
Models
· 1 src · Sep 7
Discuss
7
ByteDance Founder Zhang Yiming Personally Leads World Model Race for Spatial Video
Models
1
Sep 7
7
ByteDance Founder Zhang Yiming Personally Leads World Model Race for Spatial Video
Models
· 1 src · Sep 7
Discuss
7
GPT-6 Astra Shows 20–33% Advantage Over Rivals on Complex Cross-File Code Review
Updated Sep 14
Models
2
5d ago
7
GPT-6 Astra Shows 20–33% Advantage Over Rivals on Complex Cross-File Code Review
Upd Sep 14
Models
· 2 srcs · 5d ago
Discuss
8
Runway Launches GWM Worlds 2: Real-Time Interactive AI World Generation
Models
1
Sep 4
8
Runway Launches GWM Worlds 2: Real-Time Interactive AI World Generation
Models
· 1 src · Sep 4
Discuss
7
Ant Group Releases Ling-3.0-flash-Fin, Open-Weight 124B Finance-Specialized AI Model
Updated Sep 17
Models
2
2d ago
7
Ant Group Releases Ling-3.0-flash-Fin, Open-Weight 124B Finance-Specialized AI Model
Upd Sep 17
Models
· 2 srcs · 2d ago
Discuss
7
GPT-6 Astra Review: Autonomous Tool Use, Manager Loop, and Unreal Engine World-Building
Updated Sep 14
Models
3
5d ago
7
GPT-6 Astra Review: Autonomous Tool Use, Manager Loop, and Unreal Engine World-Building
Upd Sep 14
Models
· 3 srcs · 5d ago
Discuss
7
OpenEvidence Launches Medical AI Model Family With Darwin Achieving Perfect MedQA Score
Models
1
Sep 4
7
OpenEvidence Launches Medical AI Model Family With Darwin Achieving Perfect MedQA Score
Models
· 1 src · Sep 4
Discuss
9
OpenAI Releases GPT-6 Astra, Claims Entry Into the AGI Era
Updated Sep 19
Models
35
23h ago
9
OpenAI Releases GPT-6 Astra, Claims Entry Into the AGI Era
Top
Upd Sep 19
Models
· 35 srcs · 23h ago
Discuss
8
World Labs Launches Atlas Omni World Model With Pixel-Perfect Camera Control and 3D Reconstruction
Models
4
Sep 3
8
World Labs Launches Atlas Omni World Model With Pixel-Perfect Camera Control and 3D Reconstruction
Top
Models
· 4 srcs · Sep 3
Discuss
8
Google Releases Gemini 3.8 Flash with Frontier-Level Coding, Engineers Prefer It Over Anthropic Opus
Models
2
Sep 3
8
Google Releases Gemini 3.8 Flash with Frontier-Level Coding, Engineers Prefer It Over Anthropic Opus
Top
Models
· 2 srcs · Sep 3
Discuss
8
Meta Releases Muse Spark 1.3, Its Most Powerful LLM with 1M-Token Context and Frontier Benchmarks
Updated Sep 7
Models
9
Sep 7
8
Meta Releases Muse Spark 1.3, Its Most Powerful LLM with 1M-Token Context and Frontier Benchmarks
Top
Upd Sep 7
Models
· 9 srcs · Sep 7
Discuss
7
Saudi Humain Launches Arabic AI Model humain-m3 Built on China's MiniMax, Sparking Ally Concerns
Updated Sep 4
Models
3
Sep 4
7
Saudi Humain Launches Arabic AI Model humain-m3 Built on China's MiniMax, Sparking Ally Concerns
Upd Sep 4
Models
· 3 srcs · Sep 4
Discuss
7
Microsoft AI Launches MAI-Transcribe-2 Claiming Top Speech Recognition at $0.10/Audio Hour
Updated Sep 4
Models
4
Sep 4
7
Microsoft AI Launches MAI-Transcribe-2 Claiming Top Speech Recognition at $0.10/Audio Hour
Upd Sep 4
Models
· 4 srcs · Sep 4
Discuss
7
Claude Fable 5.1 Arrives in Cursor, Tops CursorBench 3.2 at 73.4% with Self-Verification
Models
1
Sep 3
7
Claude Fable 5.1 Arrives in Cursor, Tops CursorBench 3.2 at 73.4% with Self-Verification
Models
· 1 src · Sep 3
Discuss
7
Google DeepMind Launches WeatherNext 3, Most Accurate AI Weather Model to Date
Updated Sep 8
Models
6
Sep 8
7
Google DeepMind Launches WeatherNext 3, Most Accurate AI Weather Model to Date
Upd Sep 8
Models
· 6 srcs · Sep 8
Discuss
7
Alibaba Releases Qwen3.8-Max-0902 with Coding Post-Training, Claims Code Arena WebDev #1
Models
1
Sep 3
7
Alibaba Releases Qwen3.8-Max-0902 with Coding Post-Training, Claims Code Arena WebDev #1
Models
· 1 src · Sep 3
Discuss
6
Multiverse Computing Launches Quasar 438B, Europe's Top-Scoring AI Model
Models
1
Sep 3
6
Multiverse Computing Launches Quasar 438B, Europe's Top-Scoring AI Model
Models
· 1 src · Sep 3
Discuss
7
Google Prepares Gemini 3.8 Flash to Narrow Coding Performance Gap Against Anthropic and OpenAI
Models
1
Sep 2
7
Google Prepares Gemini 3.8 Flash to Narrow Coding Performance Gap Against Anthropic and OpenAI
Models
· 1 src · Sep 2
Discuss
9
Anthropic Launches Claude Fable 5.1 and Mythos 5.1 with 25–45% Price Cut and New Safety Tiers
Updated Sep 7
Models
11
Sep 7
9
Anthropic Launches Claude Fable 5.1 and Mythos 5.1 with 25–45% Price Cut and New Safety Tiers
Top
Upd Sep 7
Models
· 11 srcs · Sep 7
Discuss
7
Google Releases TimesFM-3 Multivariate Time Series Foundation Model
Updated Sep 1
Models
3
Sep 1
7
Google Releases TimesFM-3 Multivariate Time Series Foundation Model
Upd Sep 1
Models
· 3 srcs · Sep 1
Discuss
7
Meta Superintelligence Labs Launches Muse Voice Transcribe, First Real-Time Streaming ASR Model
Models
1
Sep 1
7
Meta Superintelligence Labs Launches Muse Voice Transcribe, First Real-Time Streaming ASR Model
Models
· 1 src · Sep 1
Discuss
Filters
Signal
Title
Category
Sources
Posted
Discuss
