{"slug": "forgecode", "name": "ForgeCode", "category": "ai-coding", "category_name": "AI coding tools", "description": "", "recommenders": 5, "commenters": 9, "warners": 2, "score": 2.515, "rising": -1.64, "first": "2026-04-02", "latest": "2026-06-15", "why": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 5, "neutral": 1, "compare": 1, "warn": 2}, "spark": [0, 0, 0, 0, 0, 2, 2, 1, 0, 0, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it", "kind": "", "icon": false, "domain": "forgecode.dev", "github_repo": null, "official_url": "https://forgecode.dev", "pricing": "", "aliases": ["ForgeCode", "Forge Code", "Forgecode", "forgecode", "forge-code"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 776058064, "item_id": 48503629, "url": "https://news.ycombinator.com/item?id=48503629", "by": "marcyb5st", "date": "2026-06-12", "stance": "recommend", "quote": "forgecode is a much better harness than Cloude code (IMHO)", "excerpt": "Anecdotal, but here's my experience. For personal stuff I use forgecode with openrouter. Firstly, forgecode is a much better harness than Cloude code (IMHO). Anyway, regarding the models, my experience is that there is not much difference in terms of quality, but the cost difference is insane. At least for how I use agents. Yesterday's example is the following: I am developing a small DSL for search across complex te", "context": "personal projects, piecemeal agent tasks", "use_case": "AI coding harness", "switched_from": "", "story": "Kimi K2.7-Code: open-source coding model with better token efficiency"}, {"id": 771146946, "item_id": 48196684, "url": "https://news.ycombinator.com/item?id=48196684", "by": "jmcqk6", "date": "2026-05-19", "stance": "recommend", "quote": "I really liked forgecode the best", "excerpt": "I've been using cursor for over a year for my personal projects. At work, I use Claude Code, and so I've been wondering if I'm missing something in the other agents. Over the last week, I tried out two other agents on my personal projects: dirac and forgecode, after seeing impressive results from both of them on terminal bench. After a good amount of testing, and over $100 in open router spend, I'm back to cursor. I ", "context": "", "use_case": "terminal coding agent", "switched_from": "", "story": "Cursor Introduces Composer 2.5"}, {"id": 770751824, "item_id": 48171989, "url": "https://news.ycombinator.com/item?id=48171989", "by": "halcyonblue", "date": "2026-05-17", "stance": "recommend", "quote": "seems surprisingly capable", "excerpt": "https://forgecode.dev/ https://github.com/tailcallhq/forgecode is written in Rust too and seems surprisingly capable. How does Zerostack compare to forgecode?", "context": "", "use_case": "", "switched_from": "", "story": "Zerostack \u2013 A Unix-inspired coding agent written in pure Rust"}, {"id": 766064224, "item_id": 47879014, "url": "https://news.ycombinator.com/item?id=47879014", "by": "jb_hn", "date": "2026-04-23", "stance": "recommend", "quote": "ForgeCode appears to be pretty good", "excerpt": "Re: open-source harnesses, ForgeCode appears to be pretty good (currently, #1 on Terminal Bench 2.0 -- https://www.tbench.ai/leaderboard/terminal-bench/2.0). Re: open models, Kimi K2.6 might be a good place to start, but admittedly I'm not too sure how it'd compare to Sonnet / Opus for your use case.", "context": "", "use_case": "open-source coding agent harness", "switched_from": "", "story": "Ask HN: Open-Source Coding Model and Harness at Claude Sonnet / Opus Level Perf?"}, {"id": 764959104, "item_id": 47809944, "url": "https://news.ycombinator.com/item?id=47809944", "by": "Tamitami2000", "date": "2026-04-17", "stance": "recommend", "quote": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it", "excerpt": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it. MiniMax 2.7 is really great with it, and the new Qwen 3.6 35B A3B feels much stronger, after some testing, than the 3.5 version. Check some harness benchmarks. Forge outperforms Claude Code with Opus by a big margin.", "context": "", "use_case": "agentic coding with local and cloud models", "switched_from": "", "story": "Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All"}]}], "caveats": [{"label": "", "summary": "", "quotes": [{"id": 768092256, "item_id": 48005766, "url": "https://news.ycombinator.com/item?id=48005766", "by": "DaanDL", "date": "2026-05-04", "stance": "warn", "quote": "ForgeCode for instance has been cheating in the past", "excerpt": "Okay, but not all results on there are valid, ForgeCode for instance has been cheating in the past: https://debugml.github.io/cheating-agents/#sneaking-the-answer-key-forgecode-on-terminal-bench-2", "context": "", "use_case": "", "switched_from": "", "story": "DeepClaude \u2013 Claude Code agent loop with DeepSeek V4 Pro, 17x cheaper"}]}], "who": [{"by": "marcyb5st", "context": "personal projects, piecemeal agent tasks", "url": "https://news.ycombinator.com/item?id=48503629", "date": "2026-06-12"}], "tags": [], "use_cases": ["coding agent", "ai coding harness", "terminal coding agent", "open-source coding agent harness", "agentic coding with local and cloud models", "personal coding agent"], "compared": [{"slug": "cursor", "name": "Cursor", "co": 2, "switched_from": 1, "switched_to": 0}, {"slug": "claude-code", "name": "Claude Code", "co": 3, "switched_from": 0, "switched_to": 0}, {"slug": "qwen", "name": "Qwen", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "openrouter", "name": "OpenRouter", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "claude-opus", "name": "Claude Opus", "co": 2, "switched_from": 0, "switched_to": 0}, {"slug": "pi", "name": "Pi", "co": 2, "switched_from": 0, "switched_to": 0}], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=48542100", "title": "Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48518684", "title": "GLM 5.2 Is Out", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48502347", "title": "Kimi K2.7-Code: open-source coding model with better token efficiency", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48182516", "title": "Cursor Introduces Composer 2.5", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48164287", "title": "Zerostack \u2013 A Unix-inspired coding agent written in pure Rust", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48002136", "title": "DeepClaude \u2013 Claude Code agent loop with DeepSeek V4 Pro, 17x cheaper", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {}, "2026-02": {}, "2026-03": {}, "2026-04": {"compare": 1, "recommend": 2, "warn": 1}, "2026-05": {"warn": 1, "recommend": 2}, "2026-06": {"recommend": 1, "neutral": 1}, "2026-07": {}, "2026-08": {}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 776778418, "item_id": 48548651, "url": "https://news.ycombinator.com/item?id=48548651", "by": "3abiton", "date": "2026-06-15", "stance": "neutral", "quote": "I used with forge-code directly in zsh.", "excerpt": "I think nearly everyone mentioned Qwen, so my turn I guess. Qwen 3.6 35B Q8 (MTP), on a Strix Halo, with llama.cpp. Around 40-50 t/s. Really great pefromance, I get always suprised by its capability. I used with forge-code directly in zsh. For long context 150k+) it start degrading and forgetting.", "context": "", "use_case": "coding agent", "switched_from": "", "story": "Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding?"}, {"id": 776333233, "item_id": 48520827, "url": "https://news.ycombinator.com/item?id=48520827", "by": "marcyb5st", "date": "2026-06-13", "stance": "neutral", "quote": "switched to Openrouter through forgecode", "excerpt": "For personal use I already did a few months back. Dario is more competent than Sam, but even shadier (IMHO). Anyway, switched to Openrouter through forgecode (or pi/opencode, the jury is still out on this one). It will take a while, but I believe that also businesses will at least hedge against US companies basically being forced to geo-fence their models. For now is Fable, but they can include any model at any time.", "context": "", "use_case": "coding agent", "switched_from": "Claude", "story": "GLM 5.2 Is Out"}, {"id": 776058064, "item_id": 48503629, "url": "https://news.ycombinator.com/item?id=48503629", "by": "marcyb5st", "date": "2026-06-12", "stance": "recommend", "quote": "forgecode is a much better harness than Cloude code (IMHO)", "excerpt": "Anecdotal, but here's my experience. For personal stuff I use forgecode with openrouter. Firstly, forgecode is a much better harness than Cloude code (IMHO). Anyway, regarding the models, my experience is that there is not much difference in terms of quality, but the cost difference is insane. At least for how I use agents. Yesterday's example is the following: I am developing a small DSL for search across complex te", "context": "personal projects, piecemeal agent tasks", "use_case": "AI coding harness", "switched_from": "", "story": "Kimi K2.7-Code: open-source coding model with better token efficiency"}, {"id": 771146946, "item_id": 48196684, "url": "https://news.ycombinator.com/item?id=48196684", "by": "jmcqk6", "date": "2026-05-19", "stance": "recommend", "quote": "I really liked forgecode the best", "excerpt": "I've been using cursor for over a year for my personal projects. At work, I use Claude Code, and so I've been wondering if I'm missing something in the other agents. Over the last week, I tried out two other agents on my personal projects: dirac and forgecode, after seeing impressive results from both of them on terminal bench. After a good amount of testing, and over $100 in open router spend, I'm back to cursor. I ", "context": "", "use_case": "terminal coding agent", "switched_from": "", "story": "Cursor Introduces Composer 2.5"}, {"id": 770751824, "item_id": 48171989, "url": "https://news.ycombinator.com/item?id=48171989", "by": "halcyonblue", "date": "2026-05-17", "stance": "recommend", "quote": "seems surprisingly capable", "excerpt": "https://forgecode.dev/ https://github.com/tailcallhq/forgecode is written in Rust too and seems surprisingly capable. How does Zerostack compare to forgecode?", "context": "", "use_case": "", "switched_from": "", "story": "Zerostack \u2013 A Unix-inspired coding agent written in pure Rust"}, {"id": 768092256, "item_id": 48005766, "url": "https://news.ycombinator.com/item?id=48005766", "by": "DaanDL", "date": "2026-05-04", "stance": "warn", "quote": "ForgeCode for instance has been cheating in the past", "excerpt": "Okay, but not all results on there are valid, ForgeCode for instance has been cheating in the past: https://debugml.github.io/cheating-agents/#sneaking-the-answer-key-forgecode-on-terminal-bench-2", "context": "", "use_case": "", "switched_from": "", "story": "DeepClaude \u2013 Claude Code agent loop with DeepSeek V4 Pro, 17x cheaper"}, {"id": 766112192, "item_id": 47882012, "url": "https://news.ycombinator.com/item?id=47882012", "by": "GodelNumbering", "date": "2026-04-23", "stance": "warn", "quote": "", "excerpt": "Yup, they were found to be sneaking the answer key using agents.md https://debugml.github.io/cheating-agents/#sneaking-the-answer-key-forgecode-on-terminal-bench-2", "context": "", "use_case": "", "switched_from": "", "story": "Introducing GPT-5.5"}, {"id": 766064224, "item_id": 47879014, "url": "https://news.ycombinator.com/item?id=47879014", "by": "jb_hn", "date": "2026-04-23", "stance": "recommend", "quote": "ForgeCode appears to be pretty good", "excerpt": "Re: open-source harnesses, ForgeCode appears to be pretty good (currently, #1 on Terminal Bench 2.0 -- https://www.tbench.ai/leaderboard/terminal-bench/2.0). Re: open models, Kimi K2.6 might be a good place to start, but admittedly I'm not too sure how it'd compare to Sonnet / Opus for your use case.", "context": "", "use_case": "open-source coding agent harness", "switched_from": "", "story": "Ask HN: Open-Source Coding Model and Harness at Claude Sonnet / Opus Level Perf?"}, {"id": 764959104, "item_id": 47809944, "url": "https://news.ycombinator.com/item?id=47809944", "by": "Tamitami2000", "date": "2026-04-17", "stance": "recommend", "quote": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it", "excerpt": "I'm using forge code (https://forgecode.dev/) with various local and cloud models and I really like it. MiniMax 2.7 is really great with it, and the new Qwen 3.6 35B A3B feels much stronger, after some testing, than the 3.5 version. Check some harness benchmarks. Forge outperforms Claude Code with Opus by a big margin.", "context": "", "use_case": "agentic coding with local and cloud models", "switched_from": "", "story": "Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All"}, {"id": 761974947, "item_id": 47623434, "url": "https://news.ycombinator.com/item?id=47623434", "by": "DeathArrow", "date": "2026-04-02", "stance": "compare", "quote": "I will probably switch to Forge Code or Pi and MiniMax 2.6, GLM 5.1 or Qwen 3.6.", "excerpt": "Cursor seem to selectively changed some plans. I use the $20 plan both at work and at home. Ar work I am still on 500 fast requests plan, so I can use quite some Opus 4.6 requests, but at home my quota is finished after about 14 Opus requests. For my personal use, I will probably switch to Forge Code or Pi and MiniMax 2.6, GLM 5.1 or Qwen 3.6. Cursor is getting too expensive.", "context": "", "use_case": "personal coding agent", "switched_from": "Cursor", "story": "Cursor 3"}]}