Long-Context LLM Inference

Local agents need memory for model execution and working history. We present JustFit, an MLX runtime that coordinates their overlapping allocations: KVExec executes and checkpoints four-bit KV with bounded workspace, PhaseSwap loads phase-dependent components, and StateTrans preserves history across execution modes. With Qwen3.8-27B MXFP4 on a 24 GiB M4 Pro MacBook, JustFit reaches 327,680 retained positions across two requests, 10.67 times the evaluated baseline's 30,720-position single-request record. One request completes the full 262,144-position native window at median 5.986 tokens/s. Each shape completes 16,384 outputs per request in three fresh processes: B1 uses one cold build and two prefix extensions; B2 uses three ordered prefix extensions (Section 4). Image encoding can proceed while preserving a live 196,608-input text request. A controlled, repetitive 32K+6K workload reaches median 18.284 tokens/s at 15,626 MiB; a separate AIME 2026 evaluation scores 29/30. Coordinating execution and state lifetimes makes longer histories feasible on personal hardware.

Long-context LLM applications such as document summarization and multi-turn agents require generation from prefixes spanning tens of thousands of tokens, making decoding latency a major bottleneck. Speculative decoding (SD) reduces latency without changing model outputs, but its speedup depends on both accepted draft tokens and draft-step latency: Lightweight drafts are fast but lack the capacity to capture long-range dependencies, whereas strong independent drafts recover acceptance but incur growing KV-access cost at long prefixes. We introduce memory-augmented drafting for long-context SD, equipping a strong independent draft with compressed draft-side KV memory: A lightweight adaptor constructs and incrementally updates this memory to retain distant information and exact recent context. The target verifier retains its full KV cache and applies the standard accept/reject rule, preserving SD's lossless guarantee. Experiments on Llama~3.1-8B and 70B targets at prefix lengths up to 32K show that our method reduces draft-side memory by over 70%. It achieves speedups of up to 2.08x and 3.33x , respectively, over autoregressive decoding.

Modern LLM systems increasingly rely on knowledge-selection processes that produce high-value structured priors, such as ranked evidence, graph topology, multimodal alignment, and confidence signals. Yet LLM serving remains fundamentally oblivious to this rich structure: once such signals are serialized into a prompt, the backend observes only a flat token sequence, forcing dense and uniform consumption of the full key-value (KV) state during decoding. We term this architectural mismatch the Knowledge Selection-Runtime Consumption (KSRC) gap: richer contexts enlarge the full-prompt KV footprint and decode-time memory traffic, increasing latency and degrading throughput even when reasoning depends on only a small fraction of the context. To bridge the gap, we propose Knowledge Access Planning (KAP), a paradigm-shifting execution abstraction that elevates structured knowledge priors from passive prompt-construction hints into first-class physical execution artifacts. KAP establishes a universal intermediate representation (IR)-the runtime access plan-which compiles structured knowledge signals to govern physical KV access without altering logical prompt semantics, model weights, or training procedures. Through this IR, KAP shifts LLM serving from token-aware context consumption to plan-driven, knowledge-aware runtime consumption. We instantiate KAP with GraphSpec, a compiler-executor realization connecting structured knowledge selection to an LLM serving backend. We derive a phase-boundary model for the positive-speedup regime of plan-guided execution. Across 4K-128K long-context QA workloads, GraphSpec maintains answer quality comparable to full-context decoding while decoupling physical KV consumption from prompt length, reducing proposal-time KV access to 5.5% of source KV state at 128K, and fundamentally shifting the scaling trajectory of long-context generation.

Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware. Existing KV cache eviction methods typically apply heuristic token scoring over all heads in GQA-based LLMs. These methods ignore the different functionalities of attention heads, leading to the eviction of critical tokens and thus degrading the performance of LLMs. To address this issue, we propose CompressKV, a resource-efficient KV-cache compression framework for GQA-based LLMs. Instead of aggregating attention scores from all heads, CompressKV identifies Semantic Retrieval Heads (SRHs) that capture both the initial and final tokens of a prompt and semantically important mid-context evidence, and uses them to select tokens whose KV pairs should be retained. Furthermore, CompressKV allocates cache budgets across layers according to offline estimates of layer-wise eviction error. Experiments on LongBench and Needle-in-a-Haystack show that CompressKV consistently outperforms existing KV-cache eviction methods across memory budgets. Notably, it preserves over 97% of full-cache performance using only 3% of the KV cache on LongBench question-answering tasks and achieves 90% accuracy with just 0.7% KV storage on Needle-in-a-Haystack. These results demonstrate an improved resource--performance trade-off for long-context LLM inference. Our code is publicly available at: https://github.com/TUDa-HWAI/CompressKV

Sparsity has long been a central theme in LLM efficiency, but its role in context processing remains unresolved. As LLM workloads shift toward longer contexts and agentic interactions, the compute and memory bottlenecks of attention become increasingly critical, raising the question of whether these constraints are fundamental. Our position is that these constraints are artificial and unnecessary, and that the future of LLM inference lies in extreme but principled sparsity along the context dimension. This position is supported by several strands of empirical and theoretical evidence. First, we find the insistence on dense attention unreasonable, since in a long context a query effectively projects O(N) attention information into a hidden space of dimension d << N, making the process inherently lossy. Second, we perform an extensive study of sparsity in LLMs spanning 20 models across five model families, varying context lengths, and different sparsity levels. We empirically demonstrate a strong trend: current LLMs, despite not being trained for context sparsity, are remarkably robust to inference-time decode sparsity across tasks of varying complexity, including retrieval, multi-hop QA, mathematical reasoning, and agentic coding. Importantly, we also show that current hardware is already sufficient to realize substantial gains from this sparsity. For example, our sparse decode kernels accelerate large-context processing by up to 10x over FlashInfer at 50x sparsity levels on hardware such as the H100. Overall, these results position extreme context sparsity not as a heuristic, but as a principled foundation for LLM inference, training, and architecture design: one that is both feasible and beneficial, and a compelling direction for future systems.

Supporting long-context LLMs is challenging due to the substantial memory demands of the key-value (KV) cache. Existing offloading systems store the full cache in host memory and selectively fetch critical entries during decoding, but this strategy quickly hits a ceiling: sparsity cannot be pushed further without degrading accuracy. As a result, when context length and batch size grow, the volume of KV transfers rises sharply and becomes the dominant source of decoding latency. We present KVDrive, a holistic multi-tier KV cache management system spanning GPU memory, host DRAM, and SSD. Unlike prior work that pursues greater sparsity through algorithmic refinements, KVDrive tackles the problem from a systems perspective - jointly orchestrating cache placement, pipeline scheduling, and cross-tier coordination to sustain high-throughput inference under tight GPU budgets. KVDrive advances three fundamental capabilities: it adapts cache management to attention behavior to maximize reuse and minimize redundant data movement; it restructures the decoding pipeline to overlap I/O- and CPU/GPU compute-bound stages, eliminating stalls across heterogeneous resources; and it harmonizes data movement across memory tiers to unlock scalable long-context inference far beyond GPU and DRAM limits. We have implemented a fully functional prototype of KVDrive and evaluated it on long-context benchmarks with popular LLMs. The system achieves up to 1.74x higher throughput compared to state-of-the-art works while preserving accuracy.

The LLM Wiki pattern, to compile and provide domain knowledge into a persistent artifact and serve it to LLMs via KV cache inference, promises context access at sub-second latency with zero retrieval failure. Realizing this requires solving the compilation gap: LLM compilation distilling raw documents into a wiki without catastrophically discarding critical facts. We characterize this gap across 17 RepLiQA domains (6,800 questions): we observe that full context KV cache inference outperforms RAG on curated knowledge (4.38 vs. 4.08 out of 5, 7.3 faster TTFT) but degrades below RAG at scale due to attention dilution, and blind compilation fails entirely (2.14 to 2.32 vs. 3.46, 53 to 60% catastrophic failure rate). To address the compilation gap, we propose WiCER (Wiki-memory Compile, Evaluate, Refine), an iterative algorithm inspired by counterexample-guided abstraction refinement (CEGAR) that closes this gap. WiCER evaluates compiled wikis against diagnostic probes, identifies dropped facts, and forces their preservation in subsequent compilations. One to two iterations recover 80% of lost quality (mean 3.24 vs. 3.47 for raw full-context across the 15 topics with baselines), reducing catastrophic failures by 55% relative. An ablation across all 17 topics confirms that targeted diagnosis (+0.95), not generic pinning (+0.16), drives the gains. All code and benchmarks are released for reproducible research.

As Large Language Models (LLMs) increasingly utilize massive context windows as working memory for autonomous tasks, their reliability fluctuates significantly depending on how information is distributed in real-world corpora. We investigate how fact placement, corpus-level distributions, and anti-hallucination ("Don't Make It Up") prompts influence model behavior by introducing a model-agnostic extended needle-in-a-haystack benchmark designed for scalability, which we apply to evaluate Gemini-2.5-flash, ChatGPT-5-mini, Claude-4.5-haiku, and Deepseek-v3.2-chat. Unlike prior work, we separately evaluate literal extraction, logical inference, and hallucination risk. We identify two critical failure modes: Distributional Collapse, where performance degrades significantly when evidence is dispersed; and a Safety Tax, where anti-hallucination prompts cause over-conservative refusal of present facts and evidence, sharply reducing accuracy. Our results suggest that many failures stem from ineffective context utilization, as models struggle to prioritize relevant information even when it is present. These findings highlight the need for model-specific robustness and effective context management to ensure reliable deployment in long-horizon agentic workflows.