Meta Muse Spark Open is live on Together AI
Meta’s Muse Spark Open, a high‑capacity US open‑weight model for agentic software engineering, is now live on Together AI as a Day 0 launch partner. The model supports a 1‑million‑token context window
Knowledge catalogue
Meta’s Muse Spark Open, a high‑capacity US open‑weight model for agentic software engineering, is now live on Together AI as a Day 0 launch partner. The model supports a 1‑million‑token context window
arXiv:2606.16710v2 Announce Type: replace-cross Abstract: Multi-agent systems (MAS) are credited with solving complex problems, yet their agents routinely act on context that may be false. Prior work
Muse Spark Open is now rolled out to Pro, Max, Pay-as-you-go, and Teams plans. Muse Spark Open is now on Ollama. @Meta's open-weight model for coding and tool use, with a 1M-token context window. Run
arXiv:2610.18612v1 Announce Type: cross Abstract: System One models such as Jev change how software makes decisions by mapping open-ended semantic contexts into structured judgments that directly driv
arXiv:2610.15259v1 Announce Type: new Abstract: Low-bit quantization reduces the cost of serving reasoning models but can substantially degrade long chains of thought. Each step reads cached keys and
arXiv:2610.19268v1 Announce Type: new Abstract: Benchmarks shape how model capabilities are measured and interpreted, yet their scores can obscure whose conventions, contexts, and judgments define suc
arXiv:2610.17639v1 Announce Type: new Abstract: Aesthetic image cropping improves composition by balancing subjects, context, and image boundaries. Large vision-language models bring semantic knowledg
arXiv:2610.17191v1 Announce Type: new Abstract: Linear-time sequence models typically store long-context information in a recurrent state. A fixed-size state limits how much information can be stored,
arXiv:2610.14207v1 Announce Type: new Abstract: Under a fixed context budget, which objects and relations should a 3D scene-grounded VLM see first? On ScanNet indoor referring benchmarks (Dai et al.,
Here is how i serve 40 people (10 of them developers) with local AI services. n8n for automatisation whisper and Vexa AI, a bot joins Teams and makes transcripts, n8n later extracts decisions and todo
arXiv:2610.13386v1 Announce Type: new Abstract: Stance detection remains challenging for large language models (LLMs) due to implicit opinions, sarcasm, and context-dependent expressions of stance. Al
arXiv:2610.13794v1 Announce Type: new Abstract: Discourse particles subtly shape the flow of discourse by reflecting conversational context and interlocutor attitudes, often leaving the truth-conditio
arXiv:2610.13334v1 Announce Type: new Abstract: Motivated by the success of large language models (LLMs) with broad generalizability and robustness to noisy or unreliable pre-training data, we explore
I've been using Claude Opus 5.5 to speed up Qwen3.8-27B on my PC (rtx 3090), it wrote a CUDA megakernel that is 1.4-1.9x faster than llama.cpp depending on the task/context length. The results and cod
arXiv:2610.11379v1 Announce Type: new Abstract: JEV models make multimodal decisions by directly scoring candidates. Although the common context is encoded once, candidate evaluation can still repeat
arXiv:2610.10795v1 Announce Type: new Abstract: We study ranking models by population-average utility from pairwise comparisons when preferences vary across users and tasks. Prior work shows that a si
In the dawn of the LLM craze, when wild Llama2's roamed free and 4k tokens was considered a fairly long context for local models, I forked an interactive fiction and MUD library called Tale to experim
arXiv:2610.11592v1 Announce Type: new Abstract: Current frontier LLMs can theoretically process long contexts with 1M tokens or more. But to what extent can they go beyond simple retrieval and perform
arXiv:2610.12005v1 Announce Type: new Abstract: Which forms of test-time compute improve the predictions of strong pretrained tabular foundation models (TFMs)? We systematically study this along three
arXiv:2610.10429v1 Announce Type: new Abstract: Autoregressive video generation requires denoising the current frames while writing their key-value representations as context for future predictions. H
arXiv:2610.09027v1 Announce Type: cross Abstract: Modern language model systems operate autonomously over increasingly long contexts containing untrusted text and images. Can an adversarial input cont
arXiv:2610.07962v1 Announce Type: cross Abstract: Where generalization capacity--the ability to extract context-invariant relational structures--first emerges remains a central question in AI and neur
arXiv:2610.07006v1 Announce Type: new Abstract: Learning effective representations helps characterize the structure and dynamics of equity markets from financial data with a low signal-to-noise ratio.
arXiv:2610.06938v1 Announce Type: new Abstract: Medical image segmentation remains fragmented along two axes: segmentation paradigms and data dimensionality. Existing methods are typically developed s
arXiv:2610.05269v1 Announce Type: new Abstract: We develop a Unified Scaling Law and a Unified Theory of Time Series Learning to understand how model capacity and historical information support foreca
Co-authors:Stenal Jolly, Strategic Cloud Engineer, GoogleAnubhav Dhawan, Software Engineer, Google Following the landmark announcement of MCP Toolbox v1.0, we're thrilled to announce that the MCP Tool
arXiv:2610.03769v1 Announce Type: new Abstract: Multi-agent LLM systems pair a sender with broad context and an executor with a limited local view. We study when a short message improves the executor'
arXiv:2610.03945v1 Announce Type: new Abstract: A constrained optimization problem may involve a parameter in its objective and active constraints, yet the final decision may remain insensitive to sma
arXiv:2610.06672v1 Announce Type: new Abstract: Long-video understanding places substantial demands on memory, as answering questions often requires retrieving information distributed across extended
Banger paper from Microsoft and colleagues. If you run agents that search a large document collection, this one is worth your time. (bookmark it) They introduce CorpusMap, which resolves recurring ent
arXiv:2610.02858v1 Announce Type: new Abstract: Language model agents are deployed with a harness, the software around the model that manages its context, tools, and feedback. When such an agent is di
arXiv:2610.03027v1 Announce Type: cross Abstract: The key-value (KV) cache becomes a major memory bottleneck in long-context LLM inference, placing substantial pressure on memory capacity and bandwidt
In our #NeurIPS2026 paper “A Minimal Interpretable Architecture for Zero-Shot Reconstruction of Dynamical Systems (DS)” (preprint: https://arxiv.org/abs/2607.14937) we reduce a DS foundation model to
I wanted to know if theres anything that can improve creative writing for our Local Models? I specificly am asking for qwen models as they are way better when it comes to researching and writing html
ggml-openvino: update to 2026.4.1, optimize performance, expand ops, improve device listing. (#29852) ggml-openvino : Qwen3.5 MoE perf (#312) Squash of ravi9#312: ggml-openvino: add detailed inference
arXiv:2610.01435v1 Announce Type: new Abstract: Tabular foundation models (TFMs) achieve strong predictive performance through in-context learning, yet repeatedly conditioning on labeled data makes in
Hi all! I am a former SWE who has recently transitioned into inference engineering. I launched a side hustle a few months back and I've just taken it full time due to excessive demand. Its been such a
arXiv:2610.00907v1 Announce Type: cross Abstract: Relational foundation models (RFMs) are pretrained once on a collection of relational databases and prediction tasks, and then applied zero-shot to pr
arXiv:2609.39131v1 Announce Type: new Abstract: Large language model (LLM) serving requires substantial memory to store model weights and KV caches. As models grow larger and contexts become longer, m
arXiv:2609.39096v1 Announce Type: new Abstract: Autoregressive video diffusion supports streaming generation and interactive control, but its KV cache grows with the generated history. Existing compre
arXiv:2609.40037v1 Announce Type: new Abstract: Few-step autoregressive video generation enables efficient streaming synthesis, but errors introduced in early temporal blocks are reused as context and
arXiv:2609.38830v1 Announce Type: new Abstract: Sparse attention is widely used to accelerate long-context inference in modern large language models (LLMs), but its input-dependent execution behavior
arXiv:2609.39001v1 Announce Type: cross Abstract: LLM services are billed per token and context windows are measured in tokens, yet the number of tokens needed for the same content varies across langu
arXiv:2609.36373v1 Announce Type: cross Abstract: A personal language agent that acts for its owner across private and shared conversations can learn a fact from one audience and later place it in the
arXiv:2510.22752v2 Announce Type: replace-cross Abstract: In-context learning is governed by both temporal and semantic relationships, shaping how Large Language Models (LLMs) retrieve contextual info
arXiv:2609.37029v1 Announce Type: cross Abstract: Speculative decoding accelerates autoregressive inference by verifying multiple draft tokens in a single target forward pass. However, as the context
I think most people are sleeping on this inference engine. I tried multiple llama.cpp forks and none of them comes close to the inference speed of Strata. Initial version had some bugs with kv cache,
arXiv:2609.36319v1 Announce Type: new Abstract: Despite the recent success of coding agents built on large language models, it remains challenging to run them over long horizons, since every observati
arXiv:2609.37082v1 Announce Type: new Abstract: Long-horizon information-seeking agents often accumulate noisy or misleading context, causing early mistakes to persist and making recovery increasingly
arXiv:2609.31640v1 Announce Type: new Abstract: Continual Learning Bench scores learning as short-horizon gain versus a reset baseline and finds naive full-context ICL strongest among the memories it
arXiv:2609.33646v1 Announce Type: cross Abstract: Browser-use agents require seamless alignment between structured web metadata and visual information, while preserving relevant context across long in
arXiv:2609.35629v1 Announce Type: new Abstract: Attention often concentrates on a small subset of tokens in the context, but which subset matters changes from one query to the next. To exploit this ch
After 237 talks across eight stages, code review emerged as the bottleneck for AI-generated code. Here are the production eval, sandbox, context, and agent experience themes that kept showing up. The
arXiv:2609.31678v1 Announce Type: new Abstract: KV cache compression is widely used for long context LLM inference under memory constraints, while deployed systems typically score refusals after gener
arXiv:2609.33889v1 Announce Type: new Abstract: At each step, decoding one sequence with a large language model rereads the projection weights, whose traffic is fixed, and the key-value (KV) cache, wh
arXiv:2609.30797v1 Announce Type: new Abstract: Text-based memory and context compression support reuse of past interactions. Resizing continuous memory changes the input to a frozen LLM, coupling cap
arXiv:2609.30878v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) provides dense teacher targets, but evaluates them only along student-sampled rollouts. When the privileged teacher f
It’s probably a bit niche and the models are a bit old at this point, but after reading about Ninfer a few months ago I did my own small vibe coded engine project for my 5080 Laptop GPU. Initially I t
Good day data nerds! I am trying to get more into running models, learning about agentic workflows, and creating my own tools. But as you do (right?) I had to fine tune my current setup. With the way
arXiv:2609.30250v1 Announce Type: new Abstract: Conspiratorial discourse on social media is not always expressed through explicit claims or stable lexical markers. The same surface content may express