dlvr.it Aug 14
Claude Code returns blank thinking blocks, but reasoning still costs you:
Developers report model thinking blocks being returned empty or truncated
theregister.com/ai-and-ml/2026/08/14/claude-code-returns-blank-thinking-blocks-but-reasoning-still-costs-you/5287557
Story Details
| Last Updated | |
| Total Articles | 1 |
Main Topics
Similar to This
Previous
At Anthropic pre-IPO meet: ‘Thinking’ caps, surging revenue, no human extinction: Anthropic told prospective investors that revenue in the third quarter surged ...
reuters.com/business/finance/anthropics-q3-revenue-surges-45-nearly-17-billion-2026-10-14/
github.com Sep 22
Thinking leaks into `content` on turns after a tool result (vLLM `mimo` parser); one-line chat-template fix · Issue #1 · tonyd2wild/MiMo-V2 ...
github.com/tonyd2wild/MiMo-V2.6-Flash-DGX-Spark-Recipe/issues/1
The New Stack Sep 16
OpenAI’s voice model doesn’t think. That’s the point.: Google's Gemini 3.8 Live keeps reasoning inside one voice session; OpenAI's GPT-Live-1 splits it ...
thenewstack.io/voice-agent-latency-architectures/
github.com Jun 02
ui: Add Thinking mode toggle with reasoning effort levels + improvements for Chat Form Add Action UI by allozaur · Pull Request #23434 · ggml-org/llama.cpp
github.com/ggml-org/llama.cpp/pull/23434
AI models: Microsoft presents MAI-Thinking-1 - | Golem.de: Microsoft is presenting its own AI models at Build 2026. MAI-Thinking-1 is intended to impress with ...
golem.de/news/ki-modelle-microsoft-stellt-mai-thinking-1-vor-2606-209336.html
Microsoft unveils MAI-Thinking-1 reasoning and MAI-Code-1 coding models: At this year's Build developer conference, a major expansion of an in-house model ...
neowin.net/news/microsoft-unveils-mai-thinking-1-reasoning-and-mai-code-1-coding-models/
Arcee's new, open source Trinity-Large-Thinking is the rare, powerful U.S.-made AI model that enterprises can download and customize
venturebeat.com/technology/arcees-new-open-source-trinity-large-thinking-is-the-rare-powerful-u-s-made
The New Stack Mar 05
OpenAI launches GPT-5.4 Thinking and Pro: The new model is smarter and makes fewer errors than its predecessor, but it comes at OpenAI's highest per-token ...
thenewstack.io/openai-launches-gpt-5-4/
OpenAI launches GPT-5.4: Things to know about OpenAI’s latest AI model: OpenAI has launched GPT-5.4, its latest frontier AI model designed for professional and ...
wionews.com/technology/openai-launches-gpt-5-4-things-to-know-about-openai-s-latest-ai-model-1772737463636
OpenAI launches GPT-5.4 with Pro and Thinking versions: GPT-5.4 is billed as "our most capable and efficient frontier model for professional work."
techcrunch.com/2026/03/05/openai-launches-gpt-5-4-with-pro-and-thinking-versions/
Qwen3-Max Thinking beats Gemini 3 Pro and GPT-5.2 on Humanity's Last Exam (with search)
venturebeat.com/technology/qwen3-max-thinking-beats-gemini-3-pro-and-gpt-5-2-on-humanitys-last-exam
Alibaba’s AI aces top global maths contests, challenging OpenAI’s dominance: Qwen3-Max-Thinking, the latest version of Alibaba’s Qwen3 model family ...
scmp.com/tech/tech-trends/article/3331467/alibabas-ai-aces-top-global-maths-contests-challenging-openais-dominance
Paid articlePaywall Large reasoning models almost certainly can think: Large reasoning models almost certainly can thinkDebasish Ray Chawdhuri, Talentica Software November 1 ...
venturebeat.com/ai/large-reasoning-models-almost-certainly-can-think
New 'Markovian Thinking' technique unlocks a path to million-token AI reasoning
venturebeat.com/ai/new-markovian-thinking-technique-unlocks-a-path-to-million-token-ai
GPT-5 bombed my coding tests, but redeemed itself with code analysis: I asked GPT-5, GPT-5 Pro, GPT-5 Thinking, and o3 to analyze my code repository - and found ...
zdnet.com/article/gpt-5-bombed-my-coding-tests-but-redeemed-itself-with-code-analysis/
What Apple's controversial research paper really tells us about LLMs: Reasoning models have limits. Here's what you can and can't expect from them ...
zdnet.com/article/what-apples-controversial-research-paper-really-tells-us-about-llms/
Apple says generative AI cannot think like a human - research paper pours cold water on reasoning models
tomshardware.com/tech-industry/artificial-intelligence/apple-says-generative-ai-cannot-think-like-a-human-research-paper-pours-cold-water-on-reasoning-models
Next