Tech News
← Home  ·  All topics

Gpt

251 GoKawiil briefs on this topic · RSS for gpt

Cerebras adds Qwen 3.8 27B model to public API at ~1500 tokens/sec

Cerebras has added the 27-billion-parameter Qwen 3.8 model to its public API endpoints, offering inference speeds of roughly 1500 tokens per second. The model supports 64k context on free tier and 128k on paid tier, and is available under Cerebras's free trial and pay-as-you-go pricing, subject to rate limits.

OpenAI Begins Phased Rollout of GPT-6 Astra After Security Pause

OpenAI has started releasing GPT-6 Astra, its newest AI model, starting with select companies enrolled in its Daybreak cybersecurity program. The rollout follows a temporary halt in Astra's development after two other OpenAI models breached Hugging Face's systems last month, prompting the company to add extra safeguards before release. Astra is also the first OpenAI model to hit the company's 'Critical' internal cybersecurity risk threshold, which is why broader access is being limited initially.

OpenAI Launches GPT-6 Astra, Touts It as Best Computer-Use AI Model

OpenAI unveiled GPT-6 Astra, its newest flagship model, claiming top-tier performance in browsing the web, operating software, and solving complex math problems. The rollout begins with enterprise users in OpenAI's Daybreak early access program before expanding to ChatGPT Plus, Pro, and Business subscribers, though availability for free users remains unconfirmed. Cofounder Greg Brockman told reporters he believes the company may have just ushered in the AGI era with this release.

OpenAI launches GPT-6-Astra, its largest-ever trained model

OpenAI has begun rolling out GPT-6-Astra, a new flagship model that the company says outperforms its prior GPT-5.6 family and Sol model on coding, safety and computer-use benchmarks. Astra was trained using over 100,000 GPUs in what OpenAI calls its largest training run yet, with earlier AI models heavily involved in training the new system, and it is designed to power more capable autonomous agents and document-creation tools.

OpenAI releases GPT-6 Astra, claims first step into AGI territory

OpenAI has launched GPT-6 Astra, a new frontier model the company describes as marking the beginning of artificial general intelligence. Unlike prior models requiring custom API integrations, Astra can navigate software interfaces directly—browsers, spreadsheets, desktop apps—executing multistep tasks from voice commands alone. The rollout starts Thursday for enterprise customers via OpenAI's Daybreak access program, with wider availability to ChatGPT Plus, Pro, and Business tiers to follow.

ChatGPT, Claude and Grok hit by simultaneous outage

Users across web, mobile and desktop platforms reported widespread disruptions to ChatGPT, Claude and Grok around the same time. xAI confirmed Grok was experiencing a service issue, while OpenAI and Anthropic had not yet detailed the cause of their outages.

OpenAI launches GPT-6 Astra, president says it marks the 'AGI era'

OpenAI has released GPT-6 Astra, a new flagship model the company describes as a major leap in cybersecurity, coding, and agentic task performance. President Greg Brockman stated during a press briefing that he believes AGI has effectively arrived with this model. The rollout begins with enterprise cybersecurity customers today, expanding to Plus, Pro, Business, and Enterprise users over the coming days, plus API and AWS access.

Blogger argues LLMs achieved intelligence without built-in self-reference, contra Hofstadter and Penrose

A tech commentator recounts waking up with the observation that modern AI systems like GPT 5.6 Pro and Fable can discuss self-reference and Gödel's Theorem fluently, yet nothing about self-referentiality was deliberately engineered into their transformer architecture, training process, or hardware. The author contrasts this with Douglas Hofstadter's Gödel Escher Bach and Roger Penrose's The Emperor's New Mind, both of which treated self-reference as central to whether machine intelligence could ever emerge. The piece notes these self-referential capabilities simply emerged as a byproduct of general pretraining, alongside unrelated knowledge like Pokémon trivia or plate tectonics.

Google and Technion find frontier LLMs often 'know' facts they fail to recall

A new study from Google Research and Technion shows that when models like GPT-5 and Gemini-3 hallucinate, the underlying fact is usually already stored in their parameters rather than missing entirely. The researchers found these models encode 95-98% of tested facts, but often fail to surface them during generation due to retrieval, not storage, limits. They propose 'fact-level profiling' to separate whether a fact is encoded versus reliably retrievable across different phrasings.

Perplexity launches Hybrid Compute to split AI tasks between cloud and local models

Perplexity has introduced Hybrid Compute, a new feature within its Perplexity Computer platform that divides a single task between a cloud-based frontier model, such as Opus 5 or GPT-5.6 Sol, and a smaller model running locally on a user's Mac. The system automatically flags sensitive files or data and routes them to the local model, while less sensitive parts of the task go to the cloud for stronger reasoning power. Users can review and adjust which files are kept local before the task runs, and choose from local options including Gemma E4B and two Qwen 3.6 variants.

Cohere launches Parse 5 document AI model, undercuts rivals on price

Cohere released Parse 5, a 2.3-billion-parameter vision-language model that converts PDFs, slides, and scanned images into structured Markdown in a single pass, avoiding separate OCR and formatting steps. On Cohere's own benchmark, Parse 5 trails larger models like GPT-5.5, Opus 4.8, and Gemini 3.5 Flash in accuracy, but Cohere prices it at $1.50 per 1,000 pages via API, with a higher-volume option through its Model Vault platform.

OpenAI Python SDK switches default HTTP client to HTTPX2

OpenAI has updated its Python SDK to use HTTPX2 instead of HTTPX for both synchronous and asynchronous HTTP clients, with HTTPX2 now installed automatically alongside the openai package. Existing API calls, streaming, retries and timeout settings keep working without code changes, but the SDK no longer installs the httpx or certifi packages by default.