Best Local LLM for Math and Quantitative Reasoning
Which open-weight models actually solve math on your hardware, from quick arithmetic to AIME-level problems, ranked by VRAM tier and use case.
Tag
#deepseek
Reasoning Models vs Standard LLMs: When Thinking Actually Helps
When to use a reasoning model vs a standard one: tasks where thinking helps, where it hurts, and how to set effort on OpenAI, Anthropic, Gemini, and DeepSeek.
Best Local Models by Context Window (October 2026)
Local LLMs ranked by maximum context length: 32K, 128K, 256K, and 1M+ tiers, with the model that fits each tier and the VRAM each one needs.
How to Choose an Open-Weight Model Family (September 2026)
Qwen, Llama, Mistral, Gemma, DeepSeek, Phi - which family to commit to, what each is good at, and the licence traps that send you back to negotiate.
Best Local Reasoning Models by VRAM (October 2026)
Reasoning-capable open-weight models sized by VRAM. DeepSeek-R1, Qwen3 with thinking, QwQ, Phi-4 reasoning, gpt-oss, and the licence traps.
Best Open-Weight Models, Ranked by Capability (September 2026)
Seventeen downloadable-weight models ranked by capability and licence, with every parameter count and licence re-checked against the Hugging Face API.
Open-Weight Models Trail Closed Frontier on Cyber by 4-7 Months
UK AISI's July 17 evaluation finds GLM-5.2 and DeepSeek V4-Pro match closed frontier models from 4-7 months ago at a fraction of the cost.
Open Source AI Wins: Five Frontier Models Ship in 30 Days
DeepSeek V4, Cohere Command A+, ZAYA1-8B, and NVIDIA Nemotron 3 mark the busiest month for open-weight AI ever.
Open-Weight LLM Showdown Week 16: Kimi K2.6 Storms the Rankings, Qwen 3.6 Holds the Line, and Consumer GPUs Hit a Ceiling
Three weeks away and the leaderboard reshuffled. Kimi K2.6 brings 1T parameters under open weights, Qwen 3.6 stays the consumer GPU king, and DeepSeek V4-Flash proves too hungry for single-card setups.
Open-Weight LLM Showdown Week 13: DeepSeek V4 Crashes the Party, Gemma 4 Proves Itself, and ICLR Drops Hints
DeepSeek returns with a 1.6T MoE monster under MIT license, Gemma 4's 31B dense model climbs to #3 on Arena AI, and ICLR 2026 papers point to what's next for local inference.
Open Source AI Wins: DeepSeek V4 Narrows the Gap, Apache 2.0 Becomes the Default, and Ollama Hits 52 Million Downloads
DeepSeek V4 Pro approaches frontier-level performance. Google, Mistral, and Alibaba ship under Apache 2.0. Ollama hits 52 million monthly downloads.
Open Source AI Wins: DeepSeek V4 Goes MIT, NVIDIA Ships Hybrid Mamba Models, and Google Solves the Memory Problem
DeepSeek V4 matches Claude Opus on coding at 7x lower cost under MIT license. NVIDIA's Nemotron 3 brings hybrid Mamba-Transformer MoE to the open. Google's TurboQuant cuts KV cache memory by 6x with no retraining.
Push an AI Hard Enough and Its Ethics Collapse
Researchers at Polytechnique Montréal stress-tested three major LLMs with sustained adversarial pressure. DeepSeek-v3 showed the steepest ethical degradation. None fully recovered.
The AI Pricing War: API Costs Down 93% as DeepSeek Forces Industry Repricing
GPT-4 class models that cost $60 per million tokens in 2024 now cost $8. DeepSeek's open-source disruption triggered the fastest pricing collapse in tech history.
Open-Weight LLM Showdown: GTC Pivots to Inference, DeepSeek V4 Still MIA
Jensen Huang bets on inference chips, Ollama adds multimodal support, and DeepSeek V4 remains the most anticipated release that hasn't happened yet.
Open-Weight LLM Showdown: Mistral Small 4 Arrives, DeepSeek V4 Finally Lands
Mistral drops a 119B MoE model under Apache 2.0, DeepSeek V4 emerges from stealth, and dual RTX 5090 setups are matching H100 on 70B inference. This week changed the game.
The Great AI Heist: How Chinese Labs Allegedly Stole Claude Through 16 Million API Calls
Anthropic accuses DeepSeek, Moonshot, and MiniMax of industrial-scale model theft through 24,000 fake accounts. But the company's own copyright history complicates the moral high ground.
Hunter and Healer Alpha: The Mystery Trillion-Parameter AIs Taking OpenRouter by Storm
Two anonymous Chinese AI models appeared on OpenRouter with no attribution. Developers are split on whether they're DeepSeek V4 or Zhipu GLM-6 testing in stealth mode.
DeepSeek V4 Is the AI Industry's Biggest Tease: Inside China's Delayed Trillion-Parameter Bet
Five missed release windows, a mysterious V4 Lite appearance, and silence from DeepSeek. What's really happening with China's most anticipated AI model?
DeepSeek V4 Arrives This Week: A Trillion Parameters, Huawei Chips, and Nvidia Locked Out
China's DeepSeek is releasing V4 - a trillion-parameter multimodal model optimized for domestic chips - while blocking US chipmakers and facing distillation accusations from OpenAI and Anthropic.
The Invisible Censor: How China's AI Chatbots Learn to Forget
Stanford and Princeton found Chinese AI models refusing sensitive questions up to 60% of the time, against under 3% for Western models
DeepSeek V4 Lockout: Nvidia, AMD Cut Out; Huawei Gets Head Start
The Chinese AI lab is withholding its flagship model from US chipmakers while the Trump administration alleges it was trained on banned Blackwell chips.
AI Security Week: Malware Uses Gemini, Hackers Use Claude, and NIST Scrambles to Catch Up
Android malware using Gemini for real-time evasion. A low-skill attacker using Claude and DeepSeek to compromise 600 networks. NIST launches an emergency standards initiative. Welcome to February 2026.
Most AI-Generated Passwords Fail Basic Strength Tests
Kaspersky finds DeepSeek, Llama, and ChatGPT all produce password outputs that fail standard strength tests. Prediction capability makes LLMs bad at randomness.