PulseAugur
EN
LIVE 07:50:22
ENTITY Conference and Labs of the Evaluation Forum

Conference and Labs of the Evaluation Forum

PulseAugur coverage of Conference and Labs of the Evaluation Forum — every cluster mentioning Conference and Labs of the Evaluation Forum across labs, papers, and developer communities, ranked by signal.

Show in brief
Total · 30d
29
43 over 90d
Releases · 30d
0
0 over 90d
Papers · 30d
6
8 over 90d
TIER MIX · 90D
TOPICS
RELATIONSHIPS
TIMELINE
  1. 2026-10-21 product_launch Cloudflare released its Clef AI model, designed for decision-making tasks. source
  2. 2026-10-19 product_launch Cloudflare launches Clef, a new tool to replace classifier calls in AI agent routers. source
  3. 2026-10-18 product_launch Cloudflare released the Clef model, a 27B parameter multimodal model. source
  4. 2026-10-17 product_launch Cloudflare has launched Clef, a new tool aimed at improving AI agent routing. source
  5. 2026-10-02 product_launch Cloudflare has released Clef, a specialized AI model for decision-making tasks. source
SENTIMENT · 30D

10 day(s) with sentiment data

RECENT · PAGE 1/3 · 43 TOTAL
  1. TOOL · CL_364272 ·

    Research questions anomaly localization in decision models

    A new research paper explores the capabilities of decision models in image anomaly detection, specifically investigating whether these models can pinpoint the location of detected anomalies. The study, using a post-trai…

  2. TOOL · CL_363011 ·

    Cloudflare launches AI models, embracing 'neolab' status

    Cloudflare is venturing into AI model development, launching two new models named Clef and Clairic. While Clef is inspired by TypeSafe.ai's Jev model, Clairic is described as a novel 'extract' model. Cloudflare's AI Pla…

  3. RESEARCH · CL_357594 ·

    Typed-decision model research shows head complexity matters less than backbone size

    Researchers have conducted a controlled study on typed-decision models, finding that the complexity of the model's 'head' (the part that scores options) does not significantly impact accuracy. Experiments showed that si…

  4. TOOL · CL_358380 ·

    Quantization noise rivals model performance shifts, study finds

    A recent study investigated the impact of quantization on decision models, revealing that factors beyond weight changes significantly affect performance. The research found that engine, batch size, and kernel selection …

  5. TOOL · CL_332974 ·

    AI models show mixed performance on rheumatology diagnostic tasks

    A new study evaluated the performance of several AI models, including Decision and System One models, on rheumatology examination questions and diagnostic case vignettes. The models were tested on two benchmarks: the Rh…

  6. TOOL · CL_330812 ·

    AI-powered DLP system prevents document leaks from image files

    This article discusses how to prevent document leaks using an AI-powered Data Loss Prevention (DLP) system for image files. It details a method that leverages Cloudflare Workers AI and Clef to achieve this.

  7. RESEARCH · CL_328345 ·

    New 'decision-only' AI models offer faster, cheaper agent loop operations · 3 sources tracked

    A new category of AI models, referred to as "System One" or decision-only models, has emerged to address the inefficiency of using large language models (LLMs) for simple decision-making tasks within agent loops. These …

  8. SIGNIFICANT · CL_314051 ·

    Cloudflare releases Clef AI for decision-making, competing with Jev

    Cloudflare has released Clef, a new type of AI model designed for decision-making rather than text generation. Unlike traditional language models, Clef takes a state and a set of typed questions, returning numerical sco…

  9. COMMENTARY · CL_309937 ·

    Decision models for AI evals: Jev, Perplexity, OpenAI, and more compared

    A new blog post explores the use of decision models for evaluation tasks, comparing several models including Jev 1.13, Perplexity Decider 1.1 27B, Liquid d1 and d1-ultra, Cloudflare Clef Flash, and OpenAI Decisions (GPT…

  10. SIGNIFICANT · CL_309827 ·

    Cloudflare releases Clairic extraction model, warns developers on output interpretation

    Cloudflare has released Clairic, a preview extraction model on its Workers AI platform, designed to identify and return specific typed fields from input text. The model, available since October 19, 2026, operates by ret…

  11. TOOL · CL_309366 ·

    AI decision APIs emerge, value shifts to structure over models

    Several companies have recently launched APIs for fast, structured decision-making, including TypeSafe, OpenAI, Cloudflare, Perplexity, and Amazon. These offerings aim to provide quicker and cheaper alternatives to trad…

  12. TOOL · CL_308505 ·

    Cloudflare's Clef aims to optimize AI agent routers with efficient, low-cost routing

    Cloudflare has introduced Clef, a new tool designed to enhance the performance of AI agent routers. Clef aims to replace the traditional classifier call within routers, which often struggles with strict output requireme…

  13. TOOL · CL_306504 ·

    Clef: A 27B multimodal model for system state interpretation

    Clef is a new 27 billion parameter multimodal model designed to interpret system states across various data types including text, JSON, images, and video. It outputs a probability for each option in a single forward pas…

  14. TOOL · CL_306077 ·

    Ollama v0.40.6 enhances Windows runners, GGUF support, and cloud features

    Ollama has released version 0.40.6, which includes several improvements. Key updates involve preventing orphaned Windows runners, adding support for q2_0 tensor and file types within GGUF, and launching the 'clef' featu…

  15. TOOL · CL_306024 ·

    Cloudflare introduces Clef for advanced AI agent routing

    Cloudflare has introduced Clef, a new tool designed to enhance AI agent routing capabilities. In benchmark tests conducted by Cloudflare, Clef achieved scores of 79.7, with another component, Jev, scoring 79.9. This dev…

  16. TOOL · CL_305034 ·

    Ollama releases v0.40.6-rc0 with updated question ID handling

    Ollama has released version v0.40.6-rc0, incorporating a pull request from Jberg that addresses question IDs within the Conference and Labs of the Evaluation Forum.

  17. TOOL · CL_299305 ·

    Four AI decision models launched in three weeks, focusing on speed and structured output

    Within a three-week period, four distinct decision models were released by different companies, none of which generate prose. These models are designed for agents and pipelines, taking a state and typed questions to ret…

  18. COMMENTARY · CL_298015 ·

    Hybrid AI decision routing shows partial success in .NET implementation

    A developer explored a hybrid AI decision-making approach, where simpler, faster models handle straightforward tasks and only uncertain decisions are escalated to more powerful, slower models. This method aims to balanc…

  19. RESEARCH · CL_296757 ·

    Prompt caching's complex tradeoffs revealed in AI performance tests

    Prompt caching, a technique aimed at optimizing AI model performance by reusing previous computations, is proving to be less straightforward than initially assumed. Recent analyses suggest that its effectiveness is high…

  20. TOOL · CL_296687 ·

    SubQ Snap launches amid decision model price war

    Subquadratic Inc. has launched SubQ Snap, a new decision model designed for tasks like routing and tool picking, capable of processing up to 384K tokens of text and images in about a second for five cents per million in…