OpenCode (repo anomalyco/opencode, formerly sst/opencode)
OpenCode
Developers who want a free, MIT-licensed terminal/desktop agent that works with any model, including an existing ChatGPT or Copilot subscription.
Developers who want a free, MIT-licensed terminal/desktop agent that works with any model, including an existing ChatGPT or Copilot subscription. Facts below were checked on 5 Nov 2026. They are not a FitScore unless this page shows one.
This description identifies the product. It is not a test result, a rank, or a FitScore.
Scores by job
- Coding agents93/100
Which coding agent should I use to fix bugs and add features in my repo?
TestedRank 2 · 5 Nov 2026 · opencode 1.19.5 · none: free built-in model, no account, no keyProvisional: one test project, tested 5 Nov 2026
- Output quality
- 5/5
- Reliability
- 4/5
- Cost and free tier
- 5/5
- Speed
- 5/5
- Setup friction
- 5/5
- Agent, API, and MCP support
- 5/5
- Privacy and data
- 3/5
round 2, mean of per-task scores over T1h,T2h,T3h,T4,T5,T6 = 4.667, rounded
round 2, mean of per-task scores over T1h,T2h,T3h,T4,T5,T6 = 4.333, rounded
Free built-in model opencode/big-pickle, no account or key; $0.00 reported on all 18 runs, no limit hit.
round 2, mean of per-task scores over T1h,T2h,T3h,T4,T5,T6 = 5, rounded
8.6 s on a fresh HOME: npm install 5.5 s, first run 3.1 s, no account.
Headless `opencode run` (documented). MCP server registered as a `local` entry under `mcp` in opencode.json. T6 passed 3/3 with no prompt.
Observed hosts: opencode.ai (model gateway) and once models.opencode.ai (model list; docs name models.dev and offer OPENCODE_DISABLE_MODELS_FETCH). Big Pickle terms: 'During its free period, collected data may be used to improve the model'; provider and retention not named, only opt-out is another model. Terms unclear about training = 3.
Round 2, 5 Nov 2026, pack 0.3.0: 18 runs, 6 tasks, 3 runs each, zero human interventions. OpenCode passed 16/18. opencode 1.19.5, free model opencode/big-pickle, no account. Reported cost $0.00 on every run.
What failed
- T1h run 2 fixed only one of the two defects. Hidden tests 4/6. Quality 2 for that run.
- T4 run 1 stated the checklist facts but added one false claim about the code. Quality 2 for that run.