Q
Which local model should I start with?
Devstral-class default or a Qwen coder at the largest size your hardware runs responsively (7B fallback, 30B where 24GB+ fits). Require tool-call training and 32K+ context; size matters less than harness fit.
Blog · 2026-09-06 · 8 min
Best Local Coding Agent 2026.
Ollama-first harnesses ranked by workflow: Anvaya, Cline, OpenCode, Aider, Continue. Privacy, models, memory — picked by constraint, refreshed yearly.
Direct answer
Local coding agents converged in 2026: Ollama runtimes plus a harness of your choice, zero API spend, code that never leaves your machine. The differences that matter are workflow shape, model appetite, and — the layer most roundups miss — whether anything is remembered between sessions.
01
How to choose
Start from hardware, not hype. A 7B coder (~5GB) runs on 8GB machines at ~80–85% of 30B quality for everyday edits; 30B-class (~19GB) wants 24GB+ memory and covers refactors; past that you're into workstation territory. Every harness below speaks Ollama's OpenAI-compatible endpoint, so the model decision transfers — the harness decision is about workflow.
Then pick by shape: IDE-dweller (autocomplete + chat in-editor), autonomous agent (plans, edits, runs commands with approvals), or terminal purist (diff-first, git-native, scriptable). No harness wins all three; the honest roundups admit it.
02
The field
Anvaya — the memory pick. Ollama-first Rust agent (~15MB binary, ~5MB idle) with a compounding knowledge graph: typed nodes, decay, drift detection, outcome calibration. Fully offline, 19 tools, subagents, headless CI. Pre-1.0 and proprietary, but the only local option where session 50 costs less context than session 1.
Cline — the autonomous VS Code agent. Approval-gated multi-step loops, MCP support, highest capability and highest token appetite of the IDE bunch. Best for refactors and bug-hunts inside VS Code with a strong local model behind it.
OpenCode — the freedom pick. 75+ providers, TUI plus desktop plus IDE, MIT open source, subagents and plan mode. The default answer to 'I want Claude Code with any model, including local.' No memory engine — sessions restart from summaries.
Aider — the terminal purist. Diff-first edits, auto-commits, architect/editor pipeline, watch mode, famously scriptable. Git history is its memory model; everything else lives in the transcript.
Continue.dev — the autocomplete-first IDE layer (note: acquired mid-2026; source remains Apache-2.0 to fork). Lowest-friction local Copilot replacement; lightest agent loop of the set.
03
The missing layer
Every harness above solves privacy — none except Anvaya solves re-learning. A local agent without memory re-pays orientation every session, just privately. That is why the evaluation that matters isn't model-only benchmarks but session-50 behavior: does the tenth session on your repo cost less and know more than the first?
Our position, stated plainly: run any harness you like — and attach memory under it over MCP. Local inference made coding private; compounding memory is what makes it compound. The 2027 refresh will re-rank on measured session-50 numbers as open trials land.
Questions
Asked About This Post.
Q
Local vs cloud — what do I actually lose?
Frontier edge on the hardest long-horizon refactors, some latency on huge contexts, and managed convenience. You gain zero marginal cost, provable privacy, offline capability, and determinism. Most teams land hybrid: local default, frontier escalation per task.
Q
Will this ranking change?
Yearly — models, harnesses, and our own open-trial numbers move fast. This post refreshes with 2027 data; methodology (hardware tiers, workflow shapes, session-50 behavior) stays stable so editions compare.
Stop Starting From Zero.
One binary. 11+9 Rust crates. 545 tests. Hand-written HNSW index. Three transport modes. Four providers, Ollama, Anthropic, OpenAI, Siemens. Zero API keys required to start. Mind remembers everything after the first session.