Section
Models – Page 2
The latest Models news, analysed daily by the i10X newsroom.
FilterCompanies & models
More stories
-
Reinforcement Learning’s Expanding Role in AI and LLMs
Reinforcement learning is shifting from robotics to powering LLMs and autonomous agents. Discover how RLHF and reward models are shaping future AI infrastructure and scientific applications.
-
LLM Optimization: Visibility vs Inference Cost Management
LLM optimization spans marketing for AI visibility and engineering for inference costs. Discover how multi-model routing cuts expenses up to 85% while maintaining quality. Explore the…
-
Local LLMs: Run AI Models Offline on Your PC
Tools like LM Studio and Ollama let you run strong local LLMs such as Llama 3.2 offline. Keep drafts private and avoid cloud risks. Discover the…
-
Local LLMs: Running Inference on Your Own GPU
Local LLMs are moving AI inference from cloud data centers to edge devices. Learn how quantization and tools like Ollama enable private, low-latency models on consumer…
-
2026 Open-Weight LLMs: Efficiency Wins Over Scale
Chinese labs are redefining open-weight LLMs in 2026 with MoE architectures that deliver massive performance using far fewer active parameters. Learn how this shifts deployment economics…
-
LLM Acquisition Collapse: Why Routers Fail to Cut Inference Costs
Learn how LLM acquisition collapse causes dynamic routers to waste inference budgets. Understand the Reward-SNR Floor and when routing policies cannot be learned from data. Explore…
-
Ollama 0.35: Jev-Style Decision Models for Local AI
Ollama 0.35 introduces native Jev-style decision models via the /v1/systemone endpoint for fast local classification and scoring. Shift from generative text to deterministic logic and explore…
-
Grok 4.7: 500K Context Window at Flat $2/$6 Pricing
xAI’s Grok 4.7 brings a 500K context window, configurable reasoning tiers, and day-one Bedrock support without raising prices. See how it enables long-running agentic tasks for…
-
DrivingBench Exposes Why Cloud LLMs Fail at Real-World Driving
DrivingBench reveals the latency and reasoning gaps when frontier models like GPT-6 Astra control a Toyota Corolla. Learn why edge computing is essential for physical AI…
-
Claude Sonnet 5.5: $2/$10 Pricing with 70.6% Terminal-Bench Score
Claude Sonnet 5.5 keeps $2/$10 rates but cuts cost-per-task 30% with adaptive thinking and 70.6% on Terminal-Bench 4.0. Discover the multi-cloud rollout and agentic workflow impact.
-
AI Translation Benchmarks Are Splintering
Traditional benchmarks like BLEU are saturating for LLMs. Explore the rise of specialized evaluations for cost, hardware constraints, low-resource languages, and edge cases in AI translation.…
-
Open-Weight Models: Ending Single-Vendor AI Lock-In
Open-weight models are reaching parity with proprietary LLMs, enabling hybrid routing architectures that cut costs and improve data control. See how enterprises are escaping vendor lock-in.
Release tracker
All labs · last 90 days · 16 releases
Gemini 4 Argon
ai.google.dev (source, opens in a new tab)Claude Haiku 5.5
support.claude.com (source, opens in a new tab)Mistral Large 4
docs.mistral.ai (source, opens in a new tab)GPT-6.1 Sol
developers.openai.com (source, opens in a new tab)Claude Sonnet 5.5
support.claude.com (source, opens in a new tab)Claude Opus 5.5
anthropic.com (source, opens in a new tab)GPT-6 Sol & Luna
developers.openai.com (source, opens in a new tab)Grok 4.7
docs.x.ai (source, opens in a new tab)
Show all 16 releases
DeepSeek-V4.1-Flash
api-docs.deepseek.com (source, opens in a new tab)GPT-6 Astra
openai.com (source, opens in a new tab)Gemini 3.8 Flash
ai.google.dev (source, opens in a new tab)Claude Fable 5.1 & Mythos 5.1
support.claude.com (source, opens in a new tab)DeepSeek-V4-Pro (GA)
api-docs.deepseek.com (source, opens in a new tab)Gemini 3.7 Flash
ai.google.dev (source, opens in a new tab)Grok 4.6
docs.x.ai (source, opens in a new tab)DeepSeek-V4-Flash
api-docs.deepseek.com (source, opens in a new tab)
Sources: company announcements & release notes.