StarCoder2 and The Stack v2

Creating a Coding Assistant with StarCoder

StarCoder: A State-of-the-Art LLM for Code

Accelerate StarCoder with 🤗 Optimum Intel on Xeon: Q8/Q4 and Speculative Decoding

StarCoder2-Instruct: Fully Transparent and Permissive Self-Alignment for Code Generation

Announcing Codestral 25.08 and the Complete Mistral Coding Stack for Enterprise

Gemini 2.5: Our most intelligent models are getting even better

Gemini 2. 5 Pro continues to be loved by developers as the best model for coding, and 2.

Causal Evidence of Stack Representations in Modeling Counter Languages Using Transformers

arXiv:2606. 03398v1 Announce Type: cross Abstract: Formal languages have proven to be effective conduits to understand the inner mechanisms of transformers.

Open R1: How to use OlympicCoder locally for coding

Training CodeParrot 🦜 from Scratch

Introducing the LiveCodeBench Leaderboard - Holistic and Contamination-Free Evaluation of Code LLMs

BigCodeArena: Judging code generations end to end with code executions

CodeTD: Topology of Attention Detects Hallucinations in Code LLMs

The paper introduces CodeTD, a novel method that uses topological data analysis of attention maps from code language models to pre‑execution assess code correctness and detect hallucinations. It quantifies prompt‑generation mismatch through topological patterns and is evaluated on multiple benchmarks (HumanEval, MBPP, BigCodeBench, MultiPL‑E) across five programming languages and ten Code LLMs up to 34B parameters. Results show CodeTD outperforms recent baselines and transfers well between coding benchmarks.