搜索“Agent”

a lengthy blog post Monday , which described Beam as a text-only mixture-of-experts model trained on high-compute reinforcement learning to be effective at reasoning, coding, and agentic tasks at “a fraction of the token cost and inference time compute” of rivals. Beam is a 501-billion-parameter model with 23 billion active parameters. It was pretrained on 23.8 trillion tokens and has a 1 million token context window. To compare, Z.ai’s GLM-5.2 has roughly 744 billion total parameters with 40 b

匹配字段正文

% of cases. Accounting and numerical precision bugs add 18%. Time validation, business rules and SSRF cover the rest. As per the model card on HF, RL rollouts ran inside the Codex agent harness on production-like software and protocol environments. Benchmark Results Cantina evaluated 60 tasks from 20 held-out vulnerability cases. Each model ran the set once, with costs estimated from provider pricing. apex-flash-1: 40/60 solved (66.7% pass@1), about $2.38; GLM-5.3-Flash (base): 36/60 solved (60.

匹配字段正文

s a question; Smart Folders: d1 files each new document into a folder, then a subfolder. It files search questions the same way, so keyword search only looks in one subfolder; Web Agent: d1 operates a flight-search website from a one-sentence goal. At each step, it picks the next action among everything the page allows; Context Compaction: d1 reads each tool output in a coding agent's session and keeps, trims, or drops it for the next task. It removes 52% of the tokens and keeps every output the

匹配字段正文

luated every model with the same eval-framework setup. In English, Kolibri leads GPQA Diamond (84.3), AIME 2025 (96.9) and AIME 2026 (96.0). It ties Qwen3.5 35B-A3B on the English agentic average at 63.4. It trails on BFCL v4, scoring 61.4 against 70.5 for Qwen3.5. The dense Qwen3.8 27B scores higher overall (80.2 EN, 79.9 DE), but activates about 8 times more parameters per token. Against its internal predecessor Kolibri Origin, Kolibri decodes about 2.7 times more text per GPU while scoring 21

匹配字段正文

ce actually supports rather than quietly broadening a study’s conclusions, and researchers need to be able to verify the final outputs. We see that in how scientists use Asta, our agentic platform for scientific work. Instead of simple keyword searches, users often bring substantial context and many constraints—for example, asking Asta to compare approaches across a body of literature while accounting for a particular method, population, or setting. Many also return to generated reports later, t

匹配字段正文

Key Points Cloudflare is releasing Clef and Clef-flash, two decision models for AI agents. Speed is the company's main selling point against frontrunner Jev. A decision model returns a brief classification with probabilities rather than a long text response. Given a customer support message, Clef assesses its urgency and identifies the team that should handle it. Downstream code can use those results to route a ticket, trigger an escalation, or hand the case to a human. According to Cloudflare,

匹配字段正文

text-to-speech models, MAI-Voice-2.1 and MAI-Voice-2.1-Flash. Artificial Analysis ranks it 1 of 38 models for final and first partial transcript accuracy. The model targets voice agents, live captions and dictation, where latency decides the experience. What Microsoft Shipped MAI-Transcribe-2-Streaming is the real-time sibling of the batch MAI-Transcribe-2, released in September. It transcribes 60 languages with automatic, continuous language detection. Audio streams in continuously, and text s

匹配字段正文

ot directly comparable. Use cases and a guardrail example The team reports early success in model routing, tool selection, argument checking, triage, guardrails, evals, and hybrid agents. In a hybrid agent, an LLM makes the hard calls and the decider handles rote ones. The repo example gates a weather tool call inside a Strands agent. A before tool call intervention asks 2 yes/no questions. Are the arguments grounded in what the user said? Is calling now premature? If the agent guessed a city, i

匹配字段正文

taBrief within range of the Claude-powered report pipeline in Asta and DR Tulu on report generation. Validating the approach Because this model was intended to work as part of our agentic Asta report generation framework (not necessarily as a standalone model), our main question was whether AstaBrief could preserve the report qualities we cared about while enabling a substantially faster and cheaper report-generation pipeline. In other words, we weren’t only asking whether the model could match

匹配字段正文

workflow evals, Clef beat Jev in 3 of 4 areas, by small margins. Invoice processing was 64.7 vs 61.8, customer service 76.3 vs 76.0, and security incidents 62.9 vs 61.7. Jev leads agent trace observability, 71.6 vs 68.5. In Cloudflare’s threat intelligence workflow, Clef classified a domain in 2.2 seconds. gpt-oss-120b took 4.7 seconds. All numbers are vendor-reported, with no independent replication yet. Feature Comparison Cloudflare’s internal Decision Index run. All 5 implement the System One

匹配字段正文

odels, and new RL fine-tuning platform | Cloudflare Blog We are introducing Clef and Clef-flash, open-source decision models hosted on Workers AI for high-speed classification and agentic workflows. https://t.co/eilE3tV0F2 BirthdayWeek— Cloudflare (@Cloudflare) October 1, 2026 ClefとClef-flashは、最近発表され話題となっているTypesafe AIのJevと同様の意思決定モデル（Decision model）で、確率を含む型付き回答を返し、意思決定が必要な際にワークフローに組み込むことができる。限定された構造化出力を低コスト、高速に一貫して生成するモデルで、AIエージェントがどのように行動すべきかを判断するのに役立つ。 ClefはQwen3.8-27Bをベースとした27Bパラメータのマルチモーダルモデ

匹配字段正文

Cohere has released Embed 5 , a new embedding model family. It targets enterprise search, RAG, and agentic retrieval. The model family ships in 2 tiers. Embed 5 Pro targets maximum retrieval quality. Embed 5 Fast targets latency and cost on the live query path. Both accept text, images, and fused text plus image inputs. Both cover 100+ languages and read up to 128K tokens. The key design choice: Pro and Fast share 1 embedding space. You can index with one and query with the other. Is it deployab

匹配字段正文

context rolls in and commits a stable transcript right away. These partials enable voice-enabled applications to act on speech before the speaker even finishes. For example, voice agents can start reasoning or calling tools mid-sentence, and live transcripts can appear as people talk. For use cases such as real-time dictation or subtitling, our internal evaluations show that words appear in the transcript 2x faster than with our closest competitor. MAI-Transcribe-2-Streaming is available at an i

匹配字段正文

n也更倾向于处理复杂且长期的深度推理任务，比如软件工程、企业专业知识梳理以及网络安全防御等。 模型的 百万Token输出 上限，就是为这种长任务准备的，它能给模型更多连续工作的余地和充足的推理深度。 以谷歌内部使用情况为例，其中Argon在谷歌各数据中心自主识别并落地内存优化方案，待方案全面部署后，预计可释放超300 TiB内存。 另一个案例中，Argon Agent围绕视频解码器的Rust移植版本，反复实验并替换了3.2万行SIMD代码，最终版本运行速度达到原Rust移植版的2.7倍。 官方还重点强调了Argon的 防御性网络安全 优势。 它能够自主发现、验证并修补20多种语言的关键软件漏洞，在Wiz内部黑箱渗透测试中，模型可以在无源代码下实时分析网络系统，高效发现攻击面、识别漏洞以及生成概念验证证据。 为了防止滥用，Argon首批进入的是谷歌的Fairwind计划，面向经过审核的网络安全防御方。 对受信任的防御团队，谷歌会放开针对网络安全任务的部分限制，Argon将帮助防守方补洞，以及用来寻找攻击入口。 不过嘴上说说得了，Argon实际跑起来怎么样，还得打个问号。 毕竟咱普罗大众还

匹配字段正文

s everything into a finished video.” See Matt’s full workflow Visit The Rundown University . How do you use AI? Tell us for a chance to be featured. 🔵 Dot - OpenAI’s new always-on agent taking on Meta’s Muse 🎨 Ideogram 4.5 - New image editing AI that holds up through repeated changes 🗣️ Mercury Voice - Inception’s cheap, speedy reasoning model for voice agents 🎬 Utopai X - MiniMax-based video AI, ranking 2 on text-to-video leaderboard Google Cloud launched Advent of Agents Season 3, featuring 31

匹配字段正文

inst GPT-6 Astra and Claude Opus 5.5. On DeepSWE it scores 77.9%, versus 74.2% for Opus 5.5 and 74.1% for Astra ( @TheRundownAI ). Internal deployments : Google reports that Argon agents freed more than 300 TiB of data-center memory and are migrating more than 800K lines of C/C++ kernel code to Rust ( @kimmonismus ). Video decoder : Agents replaced 32K lines of SIMD code with safe Rust, making the existing Rust port 2.7x faster with identical output. Research use : The team says internal agent l

匹配字段正文

onsiveness matters as much as intelligence. Powered by the Cerebras Wafer-Scale Engine, it runs at over 1,000 tokens/s, enabling near-instant feedback in live coding environments. Agentic coding has fundamentally changed software development. For the first time, machines can autonomously work for hours or days without human supervision. But this mode of interaction can also leave developers feeling out of the loop with long wait times and less opportunity to direct the work. As software developm

匹配字段正文

like GPT-5 and Claude on real workloads. High speed inference unlocks powerful new applications like instant document summarization, realtime code iteration, and low latency voice agents. These capabilities have helped Cerebras customers such as Cognition, Notion, and Mistral achieve strong differentiation in the market. Jais 2 establishes a new state of the art for Arabic language models and sets a benchmark for how a regional, culturally grounded frontier model should perform. It is available

匹配字段正文

g GLM-4.7, the latest GLM family model released from Z.ai, now available on Cerebras Inference Cloud. This model combines speed with frontier intelligence, for coding, tool-driven agents, multi-turn reasoning, and more. GLM-4.7 is a clear step up from GLM-4.6. Against leading closed models, GLM-4.7 demonstrates comparable high-quality code generation and editing, reliable tool use, and consistent multi-turn reasoning. All at up to an order of magnitude higher speed and price-performance! On benc

匹配字段正文

Tuesday, the company showed off GPT-6.1 Sol, a mere week after it launched GPT-6 Sol . OpenAI says the new model delivers nearly the same level of intelligence as GPT-6 Astra for agentic coding, computer use, and professional work, at one-fifth the standard input and output token prices. Notably, the company is not launching GPT-6.1 Astra, as was originally expected. The Wall Street Journal reported this week that OpenAI scrapped the release over safety concerns raised by researchers during int

匹配字段正文