A new study by Shanghai AI Laboratory shows that with the test-time scaling (TTS) techniques, an SLM with 1 billion parameters can outperform a 405B LLM on the complex MATH and AIME benchmarks.

Test-time scaling (TTS) is the process of giving LLMs extra compute cylces during inference to improve their performance on various tasks. Leading reasoning models, such as OpenAI o1 and DeepSeek-R1, use “internal TTS,” which means they are trained to “think” slowly by generating a long string of chain-of-thought (CoT) tokens. — Read More

As the capabilities of AI models have expanded, and as the recent paradigm of test-time compute scaling has taken off, the demand for AI inference has grown enormously. Inference revenue at major AI companies such as OpenAI and Anthropic has been growing at a rate of 3x per year or more, even as their models continue to become smaller and cheaper compared to 2023.

A few years ago, the benchmark for whether a language model was fast enough was “human reading speed”: if a model could generate 10 tokens per second when responding to a user, that was good enough. Now, as models are asked to reason at length about complex problems and are placed inside elaborate agentic loops, this benchmark has become obsolete. The benefits to serving models faster for inference are greater than ever before. Despite this, there has been little work investigating how language models can be served quickly at scale and how much we can increase their speed at the expense of paying a higher price per token.

Today, we’re releasing a model of LLM inference economics which helps answer these questions. Working with the model reveals many important facts about inference at scale that are not widely appreciated. — Read More

We’ve built a distributed cache for cloud object storage: a shared, low-latency layer that gives all compute nodes fast access to hot data.

This post looks under the hood: how hot data caching worked before, why object storage made it hard, and how the new architecture fixes it. Benchmarks included. — Read More

The rapid scaling of large language models (LLMs) has unveiled critical limitations in current hardware architectures, including constraints in memory capacity, computational efficiency, and interconnection bandwidth. DeepSeek-V3, trained on 2,048 NVIDIA H800 GPUs, demonstrates how hardware-aware model co-design can effectively address these challenges, enabling cost-efficient training and inference at scale. This paper presents an in-depth analysis of the DeepSeek-V3/R1 model architecture and its AI infrastructure, highlighting key innovations such as Multi-head Latent Attention (MLA) for enhanced memory efficiency, Mixture of Experts (MoE) architectures for optimized computation-communication trade-offs, FP8 mixed-precision training to unlock the full potential of hardware capabilities, and a Multi-Plane Network Topology to minimize cluster-level network overhead. Building on the hardware bottlenecks encountered during DeepSeek-V3’s development, we engage in a broader discussion with academic and industry peers on potential future hardware directions, including precise low-precision computation units, scale-up and scale-out convergence, and innovations in low-latency communication fabrics. These insights underscore the critical role of hardware and model co-design in meeting the escalating demands of AI workloads, offering a practical blueprint for innovation in next-generation AI systems. — Read More

#performance

Memory layers use a trainable key-value lookup mechanism to add extra parameters to a model without increasing FLOPs. Conceptually, sparsely activated memory layers complement compute-heavy dense feed-forward layers, providing dedicated capacity to store and retrieve information cheaply. This work takes memory layers beyond proof-of-concept, proving their utility at contemporary scale. On downstream tasks, language models augmented with our improved memory layer outperform dense models with more than twice the computation budget, as well as mixture-of-expert models when matched for both compute and parameters. We find gains are especially pronounced for factual tasks. We provide a fully parallelizable memory layer implementation, demonstrating scaling laws with up to 128B memory parameters, pretrained to 1 trillion tokens, comparing to base models with up to 8B parameters. — Read More

#performance

Test-time scaling is a promising new approach to language modeling that uses extra test-time compute to improve performance. Recently, OpenAI’s o1 model showed this capability but did not publicly share its methodology, leading to many replication efforts. We seek the simplest approach to achieve test-time scaling and strong reasoning performance. First, we curate a small dataset s1K of 1,000 questions paired with reasoning traces relying on three criteria we validate through ablations: difficulty, diversity, and quality. Second, we develop budget forcing to control test-time compute by forcefully terminating the model’s thinking process or lengthening it by appending “Wait” multiple times to the model’s generation when it tries to end. This can lead the model to double-check its answer, often fixing incorrect reasoning steps. After supervised finetuning the Qwen2.5-32B-Instruct language model on s1K and equipping it with budget forcing, our model s1-32B exceeds o1-preview on competition math questions by up to 27% (MATH and AIME24). Further, scaling s1-32B with budget forcing allows extrapolating beyond its performance without test-time intervention: from 50% to 57% on AIME24. Our model, data, and code are open-source at this https URL — Read More

Scaling test-time compute has emerged as a key ingredient for enabling large language models (LLMs) to solve difficult problems, but comes with high latency and inference cost. We introduce sleep-time compute, which allows models to “think” offline about contexts before queries are presented: by anticipating what queries users might ask and pre-computing useful quantities, we can significantly reduce the compute requirements at test-time. To demonstrate the efficacy of our method, we create modified versions of two reasoning tasks – Stateful GSM-Symbolic and Stateful AIME. We find that sleep-time compute can reduce the amount of test-time compute needed to achieve the same accuracy by ~ 5x on Stateful GSM-Symbolic and Stateful AIME and that by scaling sleep-time compute we can further increase accuracy by up to 13% on Stateful GSM-Symbolic and 18% on Stateful AIME. Furthermore, we introduce Multi-Query GSM-Symbolic, which extends GSM-Symbolic by including multiple related queries per context. By amortizing sleep-time compute across related queries about the same context using Multi-Query GSM-Symbolic, we can decrease the average cost per query by 2.5x. We then conduct additional analysis to understand when sleep-time compute is most effective, finding the predictability of the user query to be well correlated with the efficacy of sleep-time compute. Finally, we conduct a case-study of applying sleep-time compute to a realistic agentic SWE task. — Read More

#performance

Evaluating large language model (LLM) based applications is inherently challenging due to the unique nature of these systems. Unlike traditional software applications, where outputs are deterministic and predictable, LLMs generate outputs that can vary each time they are run, even with the same input. This variability arises from the probabilistic nature of these models, which means there is no single correct output for any given input. Consequently, testing LLM-based applications requires specialized evaluation techniques — known today as ‘evals’ — to ensure they meet performance and reliability standards. — Read More

#performance

Recently, we introduced Platinum Benchmarks as a step toward quantifying the reliability of large language models (LLMs). In that work, we revised older benchmarks to minimize label noise, such as ambiguous or mislabeled examples, and showed that frontier LLMs still make genuine errors on simple questions. For example, as part of that work we revised a 300-problem subset of GSM8K, a dataset of grade school math word problems, and found that all LLMs we tested made at least one genuine error. If certifying the precision of just a subset of the dataset can highlight new failures across models, what if we scale to all of GSM8K?

Today, we’re releasing GSM8K-Platinum, a revised version of the full GSM8K test set. Our comparative evaluation of several frontier LLMs on both the original and revised datasets demonstrates that GSM8K-Platinum provides a more accurate assessment of mathematical reasoning capabilities, revealing differences in performance that were previously hidden. — Read More

When deploying large language models (LLMs), it is important to ensure that these models are not only capable, but also reliable. Many benchmarks have been created to track LLMs’ growing capabilities, however there has been no similar focus on measuring their reliability. To understand the potential ramifications of this gap, we investigate how well current benchmarks quantify model reliability. We find that pervasive label errors can compromise these evaluations, obscuring lingering model failures and hiding unreliable behavior.

Motivated by this gap in the evaluation of reliability, we then propose the concept of so-called platinum benchmarks, i.e., benchmarks carefully curated to minimize label errors and ambiguity. As a first attempt at constructing such benchmarks, we revise examples from fifteen existing popular benchmarks. We evaluate a wide range of models on these platinum benchmarks and find that, indeed, frontier LLMs still exhibit failures on simple tasks such as elementary-level math word problems. Analyzing these failures further reveals previously unidentified patterns of problems on which frontier models consistently struggle. We provide code at this https URL — Read More