Search: “model-releases”

We analyzed DeepSeek-V4 Pro 0813 against GPT-5.6 Sol and Fable 5 on software engineering tasks using DeepSWE. DeepSeek-V4 Pro 0813 reaches 88.5% pass@4 while costing 0.24 per task, 35x less than Sol a

We think a lot about the distribution of all real-world documents, particularly for real-world use cases. We think about the distribution in terms of page length, field length, domain, and task type.

Processing img az99qopcg8jh1... So same as my prior post 1bit test... although this 1bit is a bit interesting you can read on unlsoth blog https://unsloth.ai/docs/models/qwen3.8 508 gigs being used I

A 27B agent just beat Claude Opus 4.8 and GPT-5.5 on held-out research replication. Replica turns paper replication into a scalable RL task space. Replicating a paper forces the same hypothesis-driven

arXiv:2608.12388v1 Announce Type: cross Abstract: The emergence of orientation selectivity in the primary visual cortex (V1) remains a central question in computational neuroscience. Shirazi's Bayes-M

arXiv:2608.12700v1 Announce Type: new Abstract: Systems that generate GPU kernels with language models report high correctness rates. Those rates come from a single loose test: run the kernel on a few

arXiv:2608.12960v1 Announce Type: new Abstract: Fiber tractography's ability to reconstruct the brain's structural pathways, has made it a crucial component of modern neuroimaging, enabling detailed,

arXiv:2608.12725v1 Announce Type: new Abstract: This paper addresses the challenge of multi-label defect classification in electroluminescence (EL) images of photovoltaic (PV) cells. Training models o

I'm always testing an image prompt with a picture of a historic place in my hometown – a small but well known 250,000 people town in Germany. I'll just ask the model, in which City this photo has been

arXiv:2608.12665v1 Announce Type: cross Abstract: For solving nonconvex equality-constrained optimization problems, a recent Gradient-Eigenstep Algorithm by Goyens et al.~is an iteration-efficient app

If you scroll down from the countdown at https://huggingface.co/Qwen/Qwen3.8-27B, you see a big model card with a bunch of sections: Highlights, Model Overview, Quickstart, Best Practices, Citation, e

arXiv:2602.18733v2 Announce Type: replace Abstract: Training data leakage from Large Language Models (LLMs) raises serious concerns related to privacy, security, and copyright compliance. A central ch

arXiv:2608.13329v1 Announce Type: new Abstract: A model that behaves differently when it senses it is being tested would undermine the evaluations we rely on, so recent work has sought to read that se

arXiv:2603.04243v3 Announce Type: replace Abstract: Cerebral small vessel disease (CSVD) markers, specifically enlarged perivascular spaces (EPVS) and lacunae, present a unique challenge in medical im

a very modular harness 👌 🧩 DeepSeek Harness v0.1 is now available in Developer Preview! 🔹 We’re opening it up to developers building agent harnesses worldwide and open-sourcing the codebase in MIT lic

arXiv:2608.12895v1 Announce Type: new Abstract: Compositional reliability bounds for multi-agent systems multiply component reliabilities, a step licensed by a conditional-independence assumption that

arXiv:2608.12368v1 Announce Type: new Abstract: Agreement with human judgments is a common proxy for evaluating the alignment of large language models (LLMs). Yet agreement in final labels does not sh

Anthropic PBC today revealed that it has developed an artificial intelligence model more capable than Claude Mythos 5. The company detailed the algorithm in the latest edition of its AI alignment repo

arXiv:2608.12342v1 Announce Type: new Abstract: Ensuring the accuracy of financial documents is critical for economic analysis, regulatory compliance, and corporate decision-making. Several studies ha

arXiv:2608.12522v1 Announce Type: new Abstract: LLM-based program evolution systems such as FunSearch and AlphaEvolve have shown strong ability to discover novel algorithms, but typically optimize eac

arXiv:2604.05485v2 Announce Type: replace Abstract: LLM agents call tools, query databases, delegate tasks, and trigger external side effects. Once an agent system can act in the world, the question i

arXiv:2608.13560v1 Announce Type: cross Abstract: Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process cent

dflash : clarify output logging of target_layer_ids (#27013) This commit tries to make the logging of target_layer_ids a bit clearer and easier to read. Currently the output generated looks like this:

sycl: fuse the gated-delta-net state writeback cpy (#26643) Port of #23940. Arc Pro B70, Qwen 3.6 27B Q4_K - Medium (48 of its 64 blocks run gated_delta_net), -ngl 99 -fa 1 -ctk f16 -ctv f16 -b 2048 -

ggml: force single thread on wasi (#25686) Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework Li

sycl: fuse mul_mat(gate) + mul_mat(up) + GLU for q4_K dense FFN (#26779) Measured on Arc Pro B70 (Battlemage, Level Zero), llama-bench -r 20, two interleaved rounds, tg128: qwen2.5-3B-Instruct Q4_K_M

tests : replace personal home directory paths with generic placeholders (#27043) Scrub developer-specific /home// paths from example docs and test fixtures so they don't leak into the tree. examples/t

server: allow accessing /metrics and /slots during llama_decode() (#27041) server_queue::worker call llama_decode inside yield_to_queue also handle process_mtmd_chunk clean up nits rm test Website: ht

llama : allow virtual igpu devices (#26953) llama : allow virtual igpu devices cont : better comment Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, Kleid

ggml : recurrent state rollback for ggml_ssm_scan (#26623) Initial changes for Recurrent state rollback for nemotron for cpu and cuda Removing CPU RS rollback. Will enable it in subsequent PRs additio

sync : ggml Website: https://llama.app macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (arm64, KleidiAI enabled) DISABLED macOS Intel (x64) iOS XCFramework Linux: Ubuntu x64 (CPU) Ubuntu ar

chat : pass reasoning_effort to template chat: add reasoning_effort to common_chat_templates_inputs Store OpenAI Chat Completions reasoning_effort and make it available to jinja templates (with model

jinja : fix quadratic cost in gather_string_parts (#27034) jinja : fix quadratic cost in gather_string_parts fix some comments remove test Website: https://llama.app macOS/iOS: macOS Apple Silicon (ar

arXiv:2608.12894v1 Announce Type: new Abstract: Cultural evaluation of large language models (LLMs) often focuses on high-resource standard languages, leaving regional culture and dialect communities

arXiv:2608.13069v1 Announce Type: new Abstract: Large language models (LLMs) are predominantly aligned to function as passive, sycophantic assistants. We challenge this default paradigm by empirically

arXiv:2608.12781v1 Announce Type: new Abstract: Hybrid-thinking multimodal large language models (MLLMs) allow a single model to alternate between deliberative thinking and latency-efficient non-think

arXiv:2608.13326v1 Announce Type: new Abstract: LLM benchmark scores can be precise even when the observation protocol does not identify the behavioral property they are intended to measure. In a cont

arXiv:2608.12911v1 Announce Type: new Abstract: While the privacy risks of multimodal large language models (MLLMs) have drawn significant attention, the unique vulnerabilities of domain-specific MLLM

Most Ollama frontends I've tried are chat UIs with tool calling bolted on as an afterthought, which is fine until you actually try to make a small local model do multi-step work and it silently emits

arXiv:2608.12842v1 Announce Type: new Abstract: Model merging has recently attracted significant attention as a promising paradigm for constructing unified multi-task models without requiring addition

arXiv:2608.12629v1 Announce Type: new Abstract: GPU kernel agents and GPU programming languages have advanced separately, leaving expert kernels difficult to reproduce. Agents usually treat the compil

arXiv:2608.12515v1 Announce Type: new Abstract: Assessing proxemic danger from a robot's egocentric perspective is critical for safe embodied navigation in human environments and requires both visual

arXiv:2603.14501v2 Announce Type: replace-cross Abstract: Large Language Models excel in high-resource programming languages but struggle with low-resource ones. Existing research related to low-resou

arXiv:2608.12555v1 Announce Type: new Abstract: Predictive explanation methods attribute a model output; they do not, by themselves, attribute an intervention effect on the real-world outcome. We intr

arXiv:2608.12340v1 Announce Type: new Abstract: With the rapid advancement of large language models (LLMs), generative data augmentation has attracted considerable attention for imbalanced text classi

arXiv:2608.12805v1 Announce Type: new Abstract: Access to clinical data is essential for developing reliable healthcare machine learning systems, but direct use of electronic health records is constra