Author: uet200
-
Vending-Bench 2: AI Models Put to the Test Running a Business for a Year
Key points Andon Labs released Vending-Bench 2, a long-horizon benchmark that gives each model a $500 starting balance to run a simulated vending machine business across 365 simulated days. Claude Opus 4.6 tops the leaderboard at $8,017.59 in final account balance, ahead of Claude Sonnet 4.6 at $7,204.14 and Gemini 3 Pro at $5,478.16. Andon […]
-
Qwen 3.5: Alibaba’s Native Multimodal Agent Model Arrives
Key points Alibaba’s Qwen team released Qwen 3.5 on February 16, 2026, with a flagship 397B-parameter mixture-of-experts model built for the agentic AI era. The flagship Qwen3.5-397B-A17B activates only 17 billion parameters per forward pass despite having 397 billion total parameters. Listed specifications include a 256K-token native context window, 1M tokens on the hosted Qwen3.5-Plus, […]
-
Qwen3-Coder-Next: Alibaba’s Ultra-Sparse 80B Coding Agent
Key points Alibaba’s Qwen team released Qwen3-Coder-Next in February 2026 under the Apache 2.0 license, an open-weight coding model that activates only 3 billion of its 80 billion total parameters per forward pass. On SWE-Bench Verified with the SWE-Agent scaffold the model scores 70.6%, and it also scores 62.8% on SWE-Bench Multilingual and 44.3% on […]
-
ACE-Step 1.5: Open-Source Music Generation That Rivals Commercial AI
Key points ACE-Step 1.5, released on January 28, 2026, is an open-source music generation model jointly developed by ACE Studio and StepFun. It generates a full song in under 2 seconds on an A100 GPU or under 10 seconds on an RTX 3090 while requiring less than 4GB of VRAM. The model uses a two-stage […]
-
KittenTTS: State-of-the-Art Voice Synthesis in Under 25 MB
Key points KittenTTS was released on February 19, 2026 by KittenML as an open-source text-to-speech model that fits in under 25 MB. Its smallest configuration uses 14–15 million parameters and runs entirely on CPUs without any GPU requirement. It ships in three variants: Nano at 14–15M parameters and about 25 MB INT8, Micro at 40M […]
-
Z-Image: Alibaba’s Efficient 6B Open-Source Image Generation Model
Key points Alibaba’s Tongyi MAI team released Z-Image, a family of 6-billion-parameter image generation models that it says matches closed-source models with 20B+ parameters at a fraction of the compute cost. Released in November 2025, Z-Image-Turbo ranked 1st among open-source models on the Artificial Analysis Text-to-Image Leaderboard. Z-Image is a suite of variants rather than […]
-
Voxtral Transcribe 2: Mistral’s Open Real-Time Speech-to-Text
Key points On February 4, 2026, Mistral AI launched Voxtral Transcribe 2, a dual-model family with a high-accuracy batch transcription model and an open-weights real-time model. Mistral’s batch transcription model, Voxtral Mini Transcribe V2, achieves approximately 4% word error rate on the FLEURS benchmark, outperforming GPT-4o mini, Gemini 2.5 Flash, AssemblyAI Universal and Deepgram Nova […]
-
Taalas HC1: Hardwiring Llama 3.1 Into Silicon for 17,000 Tokens/Second
Key points On February 21, 2026, Toronto-based startup Taalas unveiled the HC1, a custom ASIC that hard-codes Meta’s Llama 3.1 8B language model directly into silicon. Taalas says the HC1 delivers up to 17,000 tokens per second per user, roughly ten times faster than today’s best GPU-based solutions, at a fraction of the cost and […]
-
MiniMax M2.5: Frontier AI Performance at a Fraction of the Cost
Key points Announced on February 12, 2026, MiniMax’s M2.5 ships as open weights under a modified MIT license and, according to the company, matches leading proprietary systems at one-tenth to one-twentieth of their cost. M2.5 is a mixture-of-experts model with 229 billion total parameters, 10 billion active parameters and a 200,000-token context window, designed for […]
-
GLM-OCR: Z.ai’s 0.9B Model Takes the Top Spot on Document Understanding Benchmarks
Key points Z.ai has open-sourced GLM-OCR, a compact multimodal model with 0.9 billion parameters that ranks first on OmniDocBench V1.5 with a score of 94.62. GLM-OCR is a multimodal optical character recognition model developed by Z.ai, the commercial arm of ZhipuAI, and designed specifically for complex document understanding. It supports three core recognition modes: general […]