code llm
30 articles about code llm in AI news
Graphify Turns Codebases Into Queryable Graphs Without Vectors or LLMs
Graphify maps codebases into local knowledge graphs using tree-sitter AST parsing, no LLMs or vector stores. At 82k+ stars, it supports 40+ languages and 20+ AI assistants.
Claude Code Ships /workflows, Replaces LLM Orchestrator with Code
Claude Code /workflows replaces LLM orchestrator with code-based control flow, solving the token tax problem from multi-agent context buildup.
Claude Code's HTML Output Beats Markdown for LLM-Readable Docs
Claude Code generates HTML docs that LLMs parse more accurately than Markdown, per Thariq's analysis. Trade-off: harder for humans to edit.
Unsloth × NVIDIA Cut LLM Fine-Tuning ~25% — Three Glue-Code Wins on Blackwell
Daniel & Michael Han at Unsloth, in collaboration with NVIDIA, published a joint guide quantifying three glue-code optimizations that combine for ~25% faster LLM training on B200 Blackwell hardware. The wins target overhead around the main kernels — caching packed-sequence metadata, double-buffered gradient checkpoint reloads, and a cheaper GPT-OSS MoE router using argsort + bincount. All three are merged via public PRs.
HyEvo Framework Automates Hybrid LLM-Code Workflows, Cuts Inference Cost 19x vs. SOTA
Researchers propose HyEvo, an automated framework that generates agentic workflows combining LLM nodes for reasoning with deterministic code nodes for execution. It reduces inference cost by up to 19x and latency by 16x while outperforming existing methods on reasoning benchmarks.
How to Run Claude Code with Local LLMs Using This Open-Source Script
A new open-source script lets you connect Claude Code to local LLMs via llama.cpp, giving you full privacy and offline access.
How to Run Claude Code on Local LLMs with VibePod's New Backend Support
VibePod now lets you route Claude Code to Ollama or vLLM servers, enabling local model usage and cost savings.
How Godogen's Claude Code Skills Solve LLM Game Development
A developer built two Claude Code skills that generate complete Godot games by solving three key LLM bottlenecks: GDScript knowledge, build-time/runtime state, and visual QA.
LLM Architecture Gallery Compiles 38 Model Designs from 2024-2026 with Diagrams and Code
A new open-source repository provides annotated architecture diagrams, key design choices, and code implementations for 38 major LLMs released between 2024 and 2026, including DeepSeek V3, Qwen3 variants, and GLM-5 744B.
Open-Source Hack Enables Free Claude Code Execution with Local LLMs
Developers have discovered a method to run Anthropic's Claude Code using local LLMs without API costs or data leaving their machines. By redirecting API calls through environment variables, users can leverage open-source models like Qwen3.5 for private, cost-free coding assistance.
Code-as-Agent Harness Thesis: 88.5% Gains Without Touching the LLM
Paper shows 88.5% improvement by adapting runtime interface around frozen LLM. Harness generalizes across 18 backbones, challenging model-centric agent improvement.
Massive Activations Found in Hybrid Linear Attention LLMs
New paper finds massive activations in hybrid linear attention LLMs, forming pre-attention spikes and inter-spike plateaus. Code and checkpoints released on Hugging Face; implications for quantization and deployment.
SWE-Pruner Pro Saves 39% Tokens by Reading LLM Hidden States
SWE-Pruner Pro saves up to 39% tokens on coder LLMs by reading keep-or-prune signals from hidden states, maintaining task quality without external heuristics.
3 MCP Gateway Security Gaps LiteLLM's Audit Found (And How to Fix Them in
LiteLLM's audit revealed 3 MCP gateway gaps: fail-open resolver, unpinned servers, opt-in least-privilege. Fix them in Claude Code with version pinning and allowed_tools.
llm-anthropic 0.25 Adds Opus 4.7 with xhigh Thinking Effort — Here's How
Update to llm-anthropic 0.25 to access Claude Opus 4.7 with xhigh thinking_effort for tackling your most challenging code problems.
Developer Builds LLM Wiki 'Second Brain' for AI Coding Agents
A developer built an 'LLM Wiki' that feeds an AI coding agent's context window with a living knowledge base of a specific codebase. This aims to solve the agent's short-term memory problem, leading to more consistent and informed code generation.
QuatRoPE: New Positional Embedding Enables Linear-Scale 3D Spatial Reasoning in LLMs, Outperforming Quadratic Methods
Researchers propose QuatRoPE, a novel positional embedding method that encodes 3D object relations with linear input scaling. Paired with IGRE, it improves spatial reasoning in LLMs while preserving their original language capabilities.
OpenAI Winds Down Sora App, Reallocates Compute to Next-Gen 'Spud' LLM Development
OpenAI has completed initial development of its next major AI model, codenamed 'Spud,' and is winding down the Sora video app, which was reportedly a compute resource drain. The move reallocates critical infrastructure toward core LLM competition with Anthropic and Google.
LLM-Driven Heuristic Synthesis for Industrial Process Control: Lessons from Hot Steel Rolling
Researchers propose a framework where an LLM iteratively writes and refines human-readable Python controllers for industrial processes, using feedback from a physics simulator. The method generates auditable, verifiable code and employs a principled budget strategy, eliminating need for problem-specific tuning.
Treat Your LLM as an Untrusted Client: Hardening GitLab MCP for Enterprise Use
Secure GitLab MCP by treating the LLM as untrusted: add fail-closed Gitleaks scanning, per-project read limits, and expose only 8 essential tools to prevent data exfiltration.
LLM Gateway Moves That Cut Multi-Provider AI Bills 40–85%
Towards AI details an LLM gateway routing layer that cuts multi-provider AI costs by 40–85%, with pricing from $0.10 to $30 per million tokens. It matters for retail teams managing escalating AI spend.
Relay-OPD: On-Policy Distillation Fixes Prefix Failure in LLMs
Relay-OPD introduces on-policy distillation where a teacher briefly takes over student LLM generation to fix prefix failure, reducing error compounding without full teacher compute.
Xiaomi MiMo-V2.5 Tops Global LLM Token Usage at 31.2T/Month
Xiaomi MiMo-V2.5 leads global LLM token usage on OpenRouter at 31.2T/month, as Chinese models capture 63.5% share vs US 35.5%.
LLM Waterfall Pattern: 429 Failover Beats Retries & Circuit Breakers
The LLM waterfall pattern cascades requests across providers on 429 errors, outperforming retries and circuit breakers for zero-downtime AI inference.
100+ Papers Surveyed: LLMs' Metacognition Gap
A systematic survey of 100+ papers reveals gaps in LLM metacognition, including 10-30% miscalibration in top models like GPT-4 and Claude 3.
7 AI Agent Cost Optimization Strategies That Cut LLM Bills by Up to 90%
The source outlines seven cost optimization strategies for AI agents, including prompt compression and model routing, that can reduce LLM bills by up to 90%. This matters for retail and luxury brands deploying AI at scale where inference costs can become prohibitive.
GPT-Red: OpenAI's LLM Super-Hacker Finds 84% of Attacks, Humans 13%
OpenAI's GPT-Red LLM finds 84% of attacks vs 13% for humans, hardening GPT-5.6 Sol. Automated red-teaming shifts safety paradigm.
NVIDIA TwoTower: 2.4x Faster LLM Decoding, 98.7% Quality
NVIDIA TwoTower clones a pretrained LLM into a frozen context tower and trainable denoiser tower, achieving 2.42x faster generation with 98.7% quality on a 30B MoE model.
Alibaba's MIPI fixes LLM training-inference mismatch with direct RL
Alibaba's MIPI directly optimizes inference policy, fixing the mismatch in LLM post-training via the MIPU framework.
BayesBench: LLMs Match Bayesian Posteriors But Fail Downstream Prediction
BayesBench tests 7 LLMs on multi-turn Bayesian reasoning. Scaling improves latent inference but not prediction, exposing a critical gap for agentic deployment.