Best AI for coding
Choosing an AI for coding right now? Here's how the open options we've looked at stack up — and who each is really for. These are first-look assessments; we flag where we've gone hands-on.
Ranked by how broadly each fits real coding work — not by a benchmark number.
- 01
The most directly a coding model: a 753B open-weight flagship built for coding and agentic work, with a 1M-token context and both self-host and a hosted Z.ai API. First-look — vendor benchmarks, not yet hands-on.
Zhipu AI (Z.ai) - 02
Not a model — the open control center that RUNS coding agents (including GLM-5.2, Claude Code, Codex). Reach for it when you want autonomous, multi-step coding workflows you self-host. First-look.
All-Hands-AI - 03
A tiny 3B open reasoning model that punches above its size on code benchmarks and runs on a single GPU. The niche pick when you need something small and self-hostable for focused code reasoning. First-look.
Weibo AI
Also in this category
Claude Sonnet 5
Anthropic's Sonnet-tier model — 'the best combination of speed and intelligence.' Near-frontier on coding and agentic work at roughly a third of Opus's price, with a 1M-token context and adaptive thinking on by default.
Claude Mythos
Claude Mythos is Anthropic's cybersecurity-focused, Mythos-class model — built to autonomously find and fix software vulnerabilities. Its rollout is restricted; a safeguarded public sibling, Fable 5, carries Mythos-class capability to the broader public.
GPT-5.6
OpenAI's next flagship after GPT-5 — announced June 26, 2026 as a preview and generally available since July 9, 2026 in three tiers (Sol, Terra, Luna) with published per-token pricing.
MiniMax M3
MiniMax's third-generation flagship — M3 is an open-weight 428B-parameter MoE (~23B active per token) using MiniMax Sparse Attention (MSA), with a 1M-token context window and native multimodality (text, image, and video input). It's positioned as the first open-weight model to combine frontier coding, a 1M context, and native multimodality — and can even operate a desktop computer.