AllyshipContent Radar

Deep dives 738

Paper và bài viết kỹ thuật

Crawl live-1795068900000 · 13:22 19 thg 11, 2026 · Xuất bản 13:30 19 thg 11, 2026

Tác tử

v0.25.0

<h2><a href="https://github.com/openai/openai-agents-python/compare/v0.24.0...v0.25.0">0.25.0</a> (2026-11-17)</h2> <h2>Key Changes</h2> <p>Version 0.25.0 removes the beta Runloop sandbox provider and changes how Pydantic run contexts are checkpointed. Runloop users should preserve their workspace data before switching providers. Applications that share checkpoints across workers should upgrade readers first.</p> <h3>Highlights:</h3> <ul> <li>Migrate Runloop workloads to another supported sandbox provider, such as Modal, or remain on 0.24.0 while migrating. Provider-specific sessions and snapshots require a separate data migration.</li> <li>Resume nested agent-tool approvals with output guardrails using RunState schema <code>1.22</code>. Pydantic checkpoints now preserve default values and in-place edits. Use a custom serializer for process-local resources.</li> <li><code>OpenAIResponsesCompactionSession</code> now defaults to <code>gpt-6-luna</code>. Set <code>model="gpt-4.1"</code> to keep the previous choice. Agent model settings also follow the selected model more closely. Native Chat Completions tool requests for <code>gpt-6-sol</code> and <code>gpt-6-luna</code> default to non-reasoning mode when no effort is supplied.</li> <li>Runner handoffs can opt into recovery with <code>failure_error_function</code>. Omitting it preserves exception propagation. The demo REPL can now approve or reject individual tool calls.</li> <li>MCP v2 Streamable HTTP supports concurrent requests. MCP v1 retains serialized requests. Failed and incomplete Responses calls now retain provider-reported token usage.</li> </ul> <h3>Features</h3> <ul> <li>expose utilities for third-party session adapters (<a href="https://github.com/openai/openai-agents-python/issues/5428" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5428/hovercard">#5428</a>) (<a href="https://github.com/openai/openai-agents-python/commit/16fdb2091d0b18cc1200d26ba6a20ee6f1962455">16fdb20</a>)</li> <li><strong>sandbox:</strong> remove Runloop provider (<a href="https://github.com/openai/openai-agents-python/issues/5369" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5369/hovercard">#5369</a>) (<a href="https://github.com/openai/openai-agents-python/commit/12ce34d8438d6f1f7c964a261b89e069e8a0b3d3">12ce34d</a>)</li> </ul> <h3>Bug Fixes</h3> <ul> <li>avoid loading workspace startup files during internal sandbox operations (<a href="https://github.com/openai/openai-agents-python/pull/5406" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5406/hovercard">#5406</a>) (<a href="https://github.com/openai/openai-agents-python/commit/841dfdd938bfd4ca052475eb76d6245a356868e0">841dfdd</a>)</li> <li>preserve nested ToolContext approval decisions across resume (<a href="https://github.com/openai/openai-agents-python/pull/5633" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5633/hovercard">#5633</a>) (<a href="https://github.com/openai/openai-agents-python/commit/c200ae8bd5b3f7510f0893f467aa9d118fdc7b12">c200ae8</a>)</li> <li>avoid repeatedly fingerprinting completed tool calls (<a href="https://github.com/openai/openai-agents-python/pull/5421" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5421/hovercard">#5421</a>) (<a href="https://github.com/openai/openai-agents-python/commit/dda19f9b91cd540cedba2a2d76bcea56deabee98">dda19f9</a>)</li> <li>add opt-in handoff error recovery (<a href="https://github.com/openai/openai-agents-python/issues/5418" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5418/hovercard">#5418</a>) (<a href="https://github.com/openai/openai-agents-python/commit/a6ed84f38ed374585fb3969a8177a76755d58d94">a6ed84f</a>)</li> <li>align implicit settings with provider-selected models (<a href="https://github.com/openai/openai-agents-python/issues/5425" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5425/hovercard">#5425</a>) (<a href="https://github.com/openai/openai-agents-python/commit/a88d2919631dbd93801a79eb0f97011ef9b451b7">a88d291</a>)</li> <li>align model defaults with TS SDK (<a href="https://github.com/openai/openai-agents-python/issues/6064" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/6064/hovercard">#6064</a>) (<a href="https://github.com/openai/openai-agents-python/commit/d5287b741dfd93de3ce1625f98b342cd0c4c455e">d5287b7</a>)</li> <li>allow concurrent Streamable HTTP requests on MCP v2 (<a href="https://github.com/openai/openai-agents-python/issues/5641" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5641/hovercard">#5641</a>) (<a href="https://github.com/openai/openai-agents-python/commit/2b90566eb2adc7d94e8fd8f234b007a937a98989">2b90566</a>)</li> <li>default GPT-6 Chat Completions tool requests to non-reasoning (<a href="https://github.com/openai/openai-agents-python/issues/5420" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5420/hovercard">#5420</a>) (<a href="https://github.com/openai/openai-agents-python/commit/3396dca75663d3e72f27aed3e3a79c5d018b8332">3396dca</a>)</li> <li><strong>deps-dev:</strong> bump aiohttp from 3.14.3 to 3.14.5 (<a href="https://github.com/openai/openai-agents-python/issues/5403" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5403/hovercard">#5403</a>) (<a href="https://github.com/openai/openai-agents-python/commit/442d0ea112e62e3423d3e93478532424f71b3f8a">442d0ea</a>)</li> <li><strong>deps-dev:</strong> bump dapr from 1.18.3 to 1.19.0 (<a href="https://github.com/openai/openai-agents-python/issues/5711" data-hovercard-type="pull_request" data-hovercard-url="/openai/openai-agents-python/pull/5711/hovercard">#5711</a>) (<a href="https://github.com/openai

Nguyên văn nguồn
Chi tiết nguồn
  • v0.25.0 openai/openai-agents-python releases · 19 thg 11, 2026
Kiến trúc

Connect Amazon Bedrock AgentCore Gateway to private APIs in a hub-and-spoke network with centralized inspection

<p>In many enterprises, a central network team owns every virtual private cloud (VPC) and subnet, and every flow between VPCs must pass a central firewall in both directions. Agents that call internal REST APIs and Model Context Protocol (MCP) servers through <a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway.html" target="_blank" rel="noopener">Amazon Bedrock AgentCore Gateway</a>, a capability of Amazon Bedrock AgentCore, must meet the same two requirements. Amazon Bedrock AgentCore is a platform to build, connect, and optimize agents at scale, with any framework or model.</p> <p>This post shows a pattern that meets both requirements. The network team keeps ownership of every subnet, and traffic between the Gateway and your private APIs and MCP servers crosses the central firewall in both directions. It covers where to place the Amazon VPC Lattice resource gateway that the Gateway uses to reach private targets. It also covers who owns each piece, six configuration decisions, and the checks that confirm the pattern works.</p> <p>We built and tested the pattern in the US East (N. Virginia) AWS Region (<code>us-east-1</code>), in a lab with three member accounts and a management account in AWS Organizations. The AWS CloudFormation templates, test scripts, test evidence, and reference material (address plan, full validation checks, troubleshooting, quotas) are in the <a href="https://github.com/aws-samples/sample-bedrock-agentcore-gateway-hub-spoke-inspection" target="_blank" rel="noopener">companion repository</a>. Use the pattern as a guideline and adjust it to your own account architecture.</p> <h3 id="terms-used-in-this-post">Terms used in this post</h3> <ul> <li><strong>Resource gateway:</strong> The <a href="https://docs.aws.amazon.com/vpc-lattice/latest/ug/what-is-vpc-lattice.html" target="_blank" rel="noopener">Amazon VPC Lattice</a> entry point into a VPC: elastic network interfaces (ENIs) in subnets you choose. Backends see its addresses as the source of each connection.</li> <li><strong>Resource configuration:</strong> A VPC Lattice object that names one backend (here, an internal load balancer or VPC endpoint DNS name and port) behind a resource gateway. You can share it with other accounts through AWS Resource Access Manager (AWS RAM).</li> <li><strong>Managed mode and self-managed mode:</strong> The two <a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/vpc-egress-private-endpoints.html" target="_blank" rel="noopener">private endpoint</a> options for Gateway targets. In managed mode, AgentCore creates the resource gateway for you. In self-managed mode, you create it and give the target a resource configuration.</li> </ul> <h2 id="when-to-use-this-pattern">When to use this pattern</h2> <p>Use this pattern when all of the following are true:</p> <ul> <li>A network account owns every VPC, subnet, and AWS Transit Gateway attachment, and shares application subnets through AWS RAM.</li> <li>A central firewall in an inspection VPC must inspect traffic entering and leaving every VPC.</li> <li>Application teams manage only their own resources and security groups.</li> </ul> <p>Otherwise, one of the options in Table 1 fits better.</p> <p><em>Table 1: When another option fits better.</em></p> <table border="1px" width="100%" cellpadding="10px"> <tbody> <tr> <td><strong>Your environment</strong></td> <td><strong>Use</strong></td> </tr> <tr> <td>The Gateway account can own subnets that reach your tools</td> <td><a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/vpc-egress-private-endpoints.html#lattice-vpc-egress-managed-lattice" target="_blank" rel="noopener">Managed mode</a></td> </tr> <tr> <td>Traffic between the Gateway and the tools doesn’t need central inspection</td> <td><a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/vpc-egress-private-endpoints.html#lattice-vpc-egress-self-managed-lattice" target="_blank" rel="noopener">Self-managed mode</a>, as documented</td> </tr> <tr> <td>Tools are reachable on public endpoints</td> <td>A standard <a href="https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway.html" target="_blank" rel="noopener">Gateway target</a></td> </tr> </tbody> </table> <p>We also considered managed mode in subnets that AgentCore owns, a resource gateway in each tool VPC or in the central endpoints VPC, and internet-facing load balancers with AWS WAF per tool. The <a href="https://github.com/aws-samples/sample-bedrock-agentcore-gateway-hub-spoke-inspection/blob/main/docs/considerations.md#alternatives-we-considered" target="_blank" rel="noopener">alternatives write-up</a> explains why each falls short under these constraints.</p> <h2 id="design-requirements-for-centralized-inspection">Design requirements for centralized inspection</h2> <ul> <li>The network account owns every subnet. The Gateway and application accounts build only in subnets shared to them through AWS RAM.</li> <li>Every flow between VPCs crosses the central firewall, in both directions. The central firewall in the inspection VPC inspects the traffic between the Gateway and your private APIs and MCP servers.</li> </ul> <p>AgentCore Gateway reaches private targets through a VPC Lattice resource gateway, and the two private endpoint modes place it differently. Managed mode creates it in the Gateway account, which fits environments where that account can own the subnets it runs in. Self-managed mode, as documented, places it in subnets with direct access to the backend, which fits environments that do not require inspection between the Gateway and the tool. This environment needs both conditions at once, so the pattern uses self-managed mode with one change. The network account owns the resource gateway and places it in subnets routed through AWS Transit Gateway, so every request and reply passes the inspection VPC.</p> <h2 id="solution-overview">Solution overview</h2> <p>Give the resource gateway its

Nguyên văn nguồn
Chi tiết nguồn
Kiến trúc

From reporting to reasoning: Getting enterprise data ready for AI agents

<p><a href="https://www.linkedin.com/groups/25720044/" target="_blank" rel="noopener"><em>AWS Quick User Group DC</em></a> <em>inaugural meetup recap</em></p> <p>Many organizations are asking how to move beyond static reporting dashboards and begin using generative and agentic AI. But before an AI agent can answer questions or help teams act, it requires a trusted data foundation.</p> <div style="width: 810px" class="wp-caption alignnone"> <a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-1-2.jpeg" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-1-2.jpeg" alt="Attendees at the AWS Quick User Group DC inaugural meetup" width="800"></a> <p class="wp-caption-text">Figure 1: AWS Quick User Group DC inaugural meetup, October 2026</p> </div> <p>On Wednesday, October 7, 2026, data leaders, developers, and practitioners gathered at the <a href="https://aws.amazon.com/training/skills-centers/arlington-skills-center/" target="_blank" rel="noopener">AWS Skills Center</a> in Arlington, Virginia, for the AWS Quick User Group DC meetup. The session brought together founders, software engineers, and enterprise leaders, among them a team from Financial Independence Group (FIG). Attendees explored practical strategies for modernizing legacy architectures, preparing structured data for agentic workflows, and automating routine tasks with no-code tools.</p> <blockquote> <p><em>User groups like this provide a forum for peer-to-peer learning and collaborative problem-solving. The event offered attendees an inside look at how organizations are putting <a href="https://aws.amazon.com/quicksuite/" target="_blank" rel="noopener">AWS Quick</a> to work to drive productivity gains.</em></p> </blockquote> <h2 id="whats-new-with-aws-quick-agents-skills-and-connectors">What’s new with AWS Quick: Agents, skills, and connectors</h2> <p>For many in the room, this meetup was a first look at what agentic AI looks like in practice with AWS Quick.</p> <div style="width: 810px" class="wp-caption alignnone"> <a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-2-2.jpeg" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-2-2.jpeg" alt="Arun Ganesan presenting AWS Quick updates to the group" width="800"></a> <p class="wp-caption-text">Figure 2: Arun Ganesan presenting what’s new with AWS Quick</p> </div> <p>The meetup kicked off by exploring the latest product capabilities across AWS Quick, focusing on its evolution into a unified AI assistant for work. Rather than functioning solely as a traditional business intelligence visualization tool, AWS Quick unifies analytics, natural-language interaction, and multi-step task orchestration.</p> <ul> <li><strong>AWS Quick agents and skills</strong>: <a href="https://docs.aws.amazon.com/quicksuite/latest/userguide/working-with-agents.html" target="_blank" rel="noopener">Turn natural-language questions</a> into multi-step reusable actions that aid in decision making across enterprise data.</li> <li><strong>Third-party connectors</strong>: AWS Quick includes built-in integrations that connect tools like Salesforce, Slack, Google Drive, and Microsoft 365, breaking down data silos.</li> <li><strong>Unified workspace</strong>: A cohesive environment that bridges research, automated visual generation, and business artifact creation (presentations, documents, and spreadsheets) directly where users work.</li> </ul> <h2 id="real-world-innovation-moving-to-ai-analytics-and-agentic-systems">Real-world innovation: Moving to AI analytics and agentic systems</h2> <p>The core of the day highlighted a real-world customer journey showing how one enterprise is rethinking its data architecture so AI agents can work from governed, trusted data.</p> <h2 id="financial-independence-group-innovating-at-the-pace-of-technology">Financial Independence Group: Innovating at the pace of technology</h2> <p>Vanessa Gonzalez, Vice President of AI at Financial Independence Group (FIG), shared how FIG keeps pushing past what’s possible today. They’re adopting new tools and platforms as technology changes, and building toward AI-powered analytics and agentic systems on top of their data with AWS Quick.</p> <div style="width: 810px" class="wp-caption alignnone"> <a href="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-3-2.png" target="_blank" rel="noopener"><img src="https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/11/16/ML-22393-3-2.png" alt="Vanessa Gonzalez presenting FIG’s AI analytics journey" width="800"></a> <p class="wp-caption-text">Figure 3: Vanessa Gonzalez, Vice President of AI at FIG, presenting</p> </div> <p>Technology is changing fast, and FIG wants to be at the front of that change. Building on its existing analytics, FIG moved to AWS Quick, a platform that brings AI into its dashboards, reporting, and analytics. AWS Quick is FIG’s “single front door” for enterprise analytics. It brings together structured data such as customer relationship management (CRM) systems, databases, and finance data. It also pulls in unstructured content such as internal knowledge bases, policy documents, and collateral, through a governed data layer. That gives FIG a base for AI-powered insights and agentic workflows built on its own data.</p> <ul> <li><strong>The AI analytics loop</strong>: Moving from reporting on what already happened to a continuous cycle: Ask → Analyze → Automate → Act.</li> <li><strong>Centralized discovery</strong>: An analytics portal organized by business area (sales, operations, marketing, commissions), with built-in AI chat assistants so teams can talk directly to their documents and datasets.</li> <li><strong>Democratized insigh

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

Why your fine-tuning job takes longer than the FLOPs say

Predicting end-to-end fine-tuning time on shared GPU clusters, from 272 production runs. A preview of our upcoming paper, Predicting End-to-End Fine-Tuning Time for Generative Language Models . The full version will be on arXiv soon. GPUs are among the most expensive and scarcest resources in AI today. Anyone running a shared cluster has to decide, every day, which jobs go on which GPUs and in what order. That plan is only as good as the estimate of how long each job will take. When a job finishes hours early, GPUs sit idle until the next one is dispatched. When it runs hours late, everything queued behind it slips, experiments wait and delivery dates move.

Nguyên văn nguồn
Chi tiết nguồn
Tác tử

Notion’s Founder on the One Thing Your AI Is Missing

Câu hỏi utm_source=every&utm_medium=podcast&utm_campaign=2026-11-flagship-sponsorship&utm_content=primary Now that AI enables anyone to build their own software, should you build everything yourself?

Lovable is the fastest way to turn ideas into software people love: https://lovable.dev/grow?utm_source=every&utm_medium=podcast&utm_campaign=2026-11-flagship-sponsorship&utm_content=primary Now that AI enables anyone to build their own software, should you build everything yourself? At Thesis, Notion cofounder and CEO Ivan Zhao joins Every’s Dan Shipper to discuss which tools should and should not be vibe coded—and why you should understand how your systems work, even as agents do the building. Subscribe to Every (free for 14 days): https://every.to/youtube-every-offer?utm_source=youtube&utm_campaign=thesis2027&utm_content=thesistalktoolsforthought Links to resources mentioned in the episode: Ivan Zhao: https://x.com/ivanhzhao Ivan’s post on his 13th year at Notion, “The Next Era of Notion”: https://www.notion.com/blog/the-next-era-of-notion Notion Calendar: https://www.notion.com/product/calendar The IBM System/360, the room-sized computer Ivan references: https://www.ibm.com/history/system-360 Claude Code: https://claude.com/product/claude-code Claude: https://claude.ai ChatGPT: https://chatgpt.com Andrew Ambrosino, who leads the Codex app at OpenAI: https://ambrosino.io Timestamps: 0:00 Introduction 1:18 Why vibe coding revives the 1960s vision of software 4:40 When people should build their own tools—and when they shouldn't 7:47 Why data and context outlast UI and procedural logic 9:46 The history game Ivan is vibe coding with his wife 12:23 Bringing a 13-year-old company from pre-AI to post-AI 16:00 The TikTokification of AI and its second-order effects 19:46 Don't surrender your understanding to agents 22:01 Why builders need Lego-like blocks to vibe code 28:04 Why the next billion knowledge workers will be agents

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

b12522

<p>[CUDA] : Add SKU-aware FP16/BF16 accumulation precision gating for NV…</p>

Nguyên văn nguồn
Chi tiết nguồn
  • b12522 ggml-org/llama.cpp releases · 19 thg 11, 2026
Kỹ thuật

langchain-xai==1.4.1

<p>Changes since langchain-xai==1.4.0</p> <p>release(xai): 1.4.1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6461257539" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41931" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41931/hovercard" href="https://github.com/langchain-ai/langchain/pull/41931">#41931</a>)<br> feat(xai): send <code>langsmith_gateway</code> headers from the client (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6433425611" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41904" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41904/hovercard" href="https://github.com/langchain-ai/langchain/pull/41904">#41904</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

langchain-deepseek==1.2.1

<p>Changes since langchain-deepseek==1.2.0</p> <p>release(deepseek): 1.2.1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6461258654" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41932" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41932/hovercard" href="https://github.com/langchain-ai/langchain/pull/41932">#41932</a>)<br> feat(deepseek): send <code>langsmith_gateway</code> headers from the client (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6433426221" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41905" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41905/hovercard" href="https://github.com/langchain-ai/langchain/pull/41905">#41905</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

langchain-groq==1.2.1

<p>Changes since langchain-groq==1.2.0</p> <p>release(groq): 1.2.1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6461256462" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41930" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41930/hovercard" href="https://github.com/langchain-ai/langchain/pull/41930">#41930</a>)<br> feat(groq): send <code>langsmith_gateway</code> headers from the client (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6433424961" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41903" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41903/hovercard" href="https://github.com/langchain-ai/langchain/pull/41903">#41903</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

b12521

<details open=""> <p>webgpu: (fix CI) allow the FA mask to be broadcast over the Q batch (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6466513057" data-permission-text="Title is private" data-url="https://github.com/ggml-org/llama.cpp/issues/32535" data-hovercard-type="pull_request" data-hovercard-url="/ggml-org/llama.cpp/pull/32535/hovercard" href="https://github.com/ggml-org/llama.cpp/pull/32535">#32535</a>)</p> <ul> <li> <p>allow the mask batch to be broadcast over the Q batch</p> </li> <li> <p>use Q_ne3 for blk_batch_count of get_alloc_size function</p> </li> </ul> </details> <p><strong>Website:</strong></p> <ul> <li><a href="https://llama.app" rel="nofollow">https://llama.app</a></li> </ul> <p><strong>Attestations:</strong></p> <ul> <li><a href="https://github.com/ggml-org/llama.cpp/attestations/69252626">https://github.com/ggml-org/llama.cpp/attestations/69252626</a></li> </ul> <p><strong>macOS/iOS:</strong></p> <ul> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-macos-arm64.tar.gz">macOS Apple Silicon (arm64)</a></li> <li>macOS Apple Silicon (arm64, KleidiAI enabled) <a href="https://github.com/ggml-org/llama.cpp/pull/23780" data-hovercard-type="pull_request" data-hovercard-url="/ggml-org/llama.cpp/pull/23780/hovercard">DISABLED</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-macos-x64.tar.gz">macOS Intel (x64)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-xcframework.zip">iOS XCFramework</a></li> </ul> <p><strong>Linux:</strong></p> <ul> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-x64.tar.gz">Ubuntu x64 (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-arm64.tar.gz">Ubuntu arm64 (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-s390x.tar.gz">Ubuntu 24.04 s390x (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-26.04-s390x.tar.gz">Ubuntu 26.04 s390x (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-zdnn-s390x.tar.gz">Ubuntu 26.04 s390x (zDNN)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-vulkan-x64.tar.gz">Ubuntu x64 (Vulkan)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-vulkan-arm64.tar.gz">Ubuntu arm64 (Vulkan)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-cuda-12.8-x64.tar.gz">Ubuntu x64 (CUDA 12)</a> - <a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/cudart-llama-b12521-bin-ubuntu-cuda-12.8-x64.tar.gz">CUDA 12.8 libraries</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-cuda-13.5-x64.tar.gz">Ubuntu x64 (CUDA 13)</a> - <a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/cudart-llama-b12521-bin-ubuntu-cuda-13.5-x64.tar.gz">CUDA 13.5 libraries</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-cuda-13.5-arm64.tar.gz">Ubuntu arm64 (CUDA 13)</a> - <a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/cudart-llama-b12521-bin-ubuntu-cuda-13.5-arm64.tar.gz">CUDA 13.5 libraries</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-rocm-10.0-x64.tar.gz">Ubuntu x64 (ROCm 10.0)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-openvino-2026.4.1-x64.tar.gz">Ubuntu x64 (OpenVINO)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-sycl-fp32-x64.tar.gz">Ubuntu x64 (SYCL FP32)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-ubuntu-sycl-fp16-x64.tar.gz">Ubuntu x64 (SYCL FP16)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-linux-arm64-snapdragon.tar.gz">Linux arm64 (Snapdragon: CPU, Adreno GPU, Hexagon NPU)</a> - <a href="https://github.com/ggml-org/llama.cpp/blob/master/docs/backend/snapdragon/linux.md">setup guide</a></li> </ul> <p><strong>Android:</strong></p> <ul> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-android-arm64.tar.gz">Android arm64 (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-android-arm64-snapdragon.tar.gz">Android arm64 (Snapdragon: CPU, Adreno GPU, Hexagon NPU)</a> - <a href="https://github.com/ggml-org/llama.cpp/blob/master/docs/backend/snapdragon/README.md">setup guide</a></li> </ul> <p><strong>Windows:</strong></p> <ul> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-win-cpu-x64.zip">Windows x64 (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-win-cpu-arm64.zip">Windows arm64 (CPU)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-win-opencl-adreno-arm64.zip">Windows arm64 (OpenCL Adreno)</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-win-cuda-12.4-x64.zip">Windows x64 (CUDA 12)</a> - <a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/cudart-llama-bin-win-cuda-12.4-x64.zip">CUDA 12.4 DLLs</a></li> <li><a href="https://github.com/ggml-org/llama.cpp/releases/download/b12521/llama-b12521-bin-win-cuda-13.5-x64.zip">Windows x64 (CUDA 13)</a> - <a href="https://github.com/ggml-org/llama.cpp/releases/download/b1

Nguyên văn nguồn
Chi tiết nguồn
  • b12521 ggml-org/llama.cpp releases · 19 thg 11, 2026
Kỹ thuật

datasette-arrows 0.1a0

<p><strong>Release:</strong> <a href="https://github.com/datasette/datasette-arrows/releases/tag/0.1a0">datasette-arrows 0.1a0</a></p> <p><a href="https://docs.datasette.io/en/latest/changelog.html#a43-2026-11-15">Datasette 1.0a43</a> added the ability for plugins to implement streaming export formats. Datasette core supports streaming CSV and TSV - you can export millions of lines of data from a table without needing to load it all into memory, because those two file formats support streaming data in chunks. The <a href="https://docs.datasette.io/en/latest/plugin_hooks.html#register-output-renderer-datasette">register_output_renderer()</a> plugin hook can now be used to implement the same pattern for other formats.</p> <p>I've learned never to release a new plugin feature without accompanying it with a plugin that uses it.</p> <p>Claude Fable 5.2 suggested the <a href="https://arrow.apache.org/docs/python/ipc.html">Apache Arrow streaming</a> format as a good option, and built this plugin in almost a single shot.</p> <p>Add the plugin to Datasette (<code>datasette install datasette-arrows</code>) and the export dialog gains this new Arrow option:</p> <p><img alt="Export data modal for the threads table. The TSV and CSV exports are now joined by Arrow, showing &quot;All 927 rows&quot; and the text Apache Arrow IPC stream, for DuckDB, pandas, Polars and other tools. There is a Download button and a Copy URL button." src="https://static.simonwillison.net/static/2026/arrow-export-dialog.webp" /></p> <p>This screenshot is from Datasette's <code>main</code> branch - the most recent release doesn't yet group streaming formats such as Arrow with the TSV and CSV in the "All rows in this table" section, but that's coming in the next alpha release.</p> <p>Tags: <a href="https://simonwillison.net/tags/datasette">datasette</a>, <a href="https://simonwillison.net/tags/projects">projects</a></p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

The Gemini app is now available on lower-memory Android devices.

<img src="https://storage.googleapis.com/gweb-uniblog-publish-prod/images/Gemini_app_Android_social_share.max-600x600.format-webp.webp">The Google Gemini app now supports lower-memory Android devices. Access Gemini Live and custom image creation on your phone. Download the app today.

Nguyên văn nguồn
Chi tiết nguồn
Đánh giá

Build Training-Ready Physical AI Datasets with Two Prompts in NVIDIA Cosmos

World model accuracy starts with high-quality, diverse training datasets. Yet capturing rare defects, nighttime incidents, and other challenging scenarios is... World model accuracy starts with high-quality, diverse training datasets. Yet capturing rare defects, nighttime incidents, and other challenging scenarios is costly and time-consuming. Turning raw data into training-ready datasets adds complexity, particularly when using disconnected tools and manual processes for curation, labeling, generation, and evaluation. NVIDIA technology is helping… Source

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

langchain-openai==1.8.2

<p>Changes since langchain-openai==1.8.1</p> <p>fix(openai): don't share the async HTTP client between models (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6470754055" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41944" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41944/hovercard" href="https://github.com/langchain-ai/langchain/pull/41944">#41944</a>)<br> feat(openai): support GPT-6.2 Astra and Sol (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6451398417" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41922" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41922/hovercard" href="https://github.com/langchain-ai/langchain/pull/41922">#41922</a>)<br> feat(openai): send <code>langsmith_gateway</code> headers from the client (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6433423744" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41901" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41901/hovercard" href="https://github.com/langchain-ai/langchain/pull/41901">#41901</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

langchain==1.5.1

<p>Changes since langchain==1.5.0</p> <p>release(langchain): 1.5.1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6470665573" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41943" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41943/hovercard" href="https://github.com/langchain-ai/langchain/pull/41943">#41943</a>)<br> fix(core): keep warning suppression context-local (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6408599282" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41881" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41881/hovercard" href="https://github.com/langchain-ai/langchain/pull/41881">#41881</a>)<br> fix(langchain): avoid warning-filter races in provider inference (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6408542790" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41880" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41880/hovercard" href="https://github.com/langchain-ai/langchain/pull/41880">#41880</a>)<br> fix(langchain): avoid warning-filter races in <code>ModelRequest</code> (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6408152317" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41879" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41879/hovercard" href="https://github.com/langchain-ai/langchain/pull/41879">#41879</a>)<br> chore(deps): bump opentelemetry-api from 1.45.1 to 1.45.2 in /libs/langchain_v1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6377748806" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41845" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41845/hovercard" href="https://github.com/langchain-ai/langchain/pull/41845">#41845</a>)<br> chore(deps): bump dnspython from 2.9.1 to 2.9.2 in /libs/langchain_v1 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6379714412" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41849" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41849/hovercard" href="https://github.com/langchain-ai/langchain/pull/41849">#41849</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

b12520

<p>ggml-cpu : use all threads for large GET_ROWS (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6459250292" data-permission-text="Title is private" data-url="https://github.com/ggml-org/llama.cpp/issues/32514" data-hovercard-type="pull_request" data-hovercard-url="/ggml-org/llama.cpp/pull/32514/hovercard" href="https://github.com/ggml-org/llama.cpp/pull/32514">#32514</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
  • b12520 ggml-org/llama.cpp releases · 18 thg 11, 2026
Kỹ thuật

langchain-core==1.8.3

<p>Changes since langchain-core==1.8.2</p> <p>release(core): 1.8.3 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6470565112" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41942" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41942/hovercard" href="https://github.com/langchain-ai/langchain/pull/41942">#41942</a>)<br> fix(core): keep warning suppression context-local (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6408599282" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41881" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41881/hovercard" href="https://github.com/langchain-ai/langchain/pull/41881">#41881</a>)<br> chore(infra): update vulnerable development dependencies (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6461990506" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41933" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41933/hovercard" href="https://github.com/langchain-ai/langchain/pull/41933">#41933</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
Kỹ thuật

langchain-anthropic==1.7.11

<p>Changes since langchain-anthropic==1.7.10</p> <p>release(anthropic): 1.7.11 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6461255356" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41929" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41929/hovercard" href="https://github.com/langchain-ai/langchain/pull/41929">#41929</a>)<br> feat(anthropic): send <code>langsmith_gateway</code> headers from the client (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6433424287" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41902" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41902/hovercard" href="https://github.com/langchain-ai/langchain/pull/41902">#41902</a>)<br> feat(anthropic): support Claude Fable and Mythos 5.2 (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6293362193" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41770" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41770/hovercard" href="https://github.com/langchain-ai/langchain/pull/41770">#41770</a>)<br> fix(anthropic): emit a chunk when an empty thinking block starts (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6434857275" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41907" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41907/hovercard" href="https://github.com/langchain-ai/langchain/pull/41907">#41907</a>)<br> feat(partners): Remove EOL'd Python 3.10 from partner packages (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="5886024423" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41268" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41268/hovercard" href="https://github.com/langchain-ai/langchain/pull/41268">#41268</a>)<br> chore(model-profiles): refresh model profile data (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6324372002" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41791" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41791/hovercard" href="https://github.com/langchain-ai/langchain/pull/41791">#41791</a>)<br> chore(model-profiles): refresh model profile data (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6302905023" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41781" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41781/hovercard" href="https://github.com/langchain-ai/langchain/pull/41781">#41781</a>)<br> fix(anthropic): raise <code>ModelRefusalError</code> for refusals when tool calls are not forced (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6226735008" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41715" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41715/hovercard" href="https://github.com/langchain-ai/langchain/pull/41715">#41715</a>)<br> fix(anthropic): preserve nested <code>caller</code> on server calls and results (<a class="issue-link js-issue-link" data-error-text="Failed to load title" data-id="6205834378" data-permission-text="Title is private" data-url="https://github.com/langchain-ai/langchain/issues/41678" data-hovercard-type="pull_request" data-hovercard-url="/langchain-ai/langchain/pull/41678/hovercard" href="https://github.com/langchain-ai/langchain/pull/41678">#41678</a>)</p>

Nguyên văn nguồn
Chi tiết nguồn
221–240 / 562
Trang 12 / 29