41554 blogs · [ { "id": "01a087e6-0358-734e-9f1a-37a8a292d1f4", "title": "On 25 Stet Tasks, Opus 4.8 Wrote Smaller Patches. Opus 5 Searched Wider.", "url": "https://www.stet.sh/blog/opus-4-8-vs-opus-5-same-score-different-routes", "published_at": "2026-08-25T12:00:00+00:00" }, { "id": "01a087e6-0358-734e-9f1a-37a8a38ed042", "title": "I Tested Six Ways to Save $$$ on 5.6 Sol. The only one that worked was using Terra", "url": "https://www.stet.sh/blog/gpt-56-token-saving-modes", "published_at": "2026-07-20T12:00:00+00:00" }, { "id": "01a087e6-0358-734e-9f1a-37a8a4781825", "title": "Sonnet 5 vs Opus 4.8: how they behave and when to use each", "url": "https://www.stet.sh/blog/sonnet-5-vs-opus-4-8-reasoning-dial", "published_at": "2026-07-08T12:00:00+00:00" }, { "id": "01a087e6-0358-734e-9f1a-37a8a4d486f1", "title": "GLM 5.2 on 50 real Go and Rust PRs: last on quality, and not the cheapest", "url": "https://www.stet.sh/blog/glm-5-2-passes-tests-fails-review", "published_at": "2026-06-19T12:00:00+00:00" }, { "id": "01a087e6-0358-734e-9f1a-37a8a5703513", "title": "Composer 2.5 was 6.5x cheaper on 50 real Rust and Go PRs. Here's where it's actually safe.", "url": "https://www.stet.sh/blog/composer-25-cheaper-where-its-safe", "published_at": "2026-06-18T12:00:00+00:00" }, { "id": "01a087e6-0358-734e-9f1a-37a8a585691b", "title": "When Fable 5 Is Worth the Premium", "url": "https://www.stet.sh/blog/when-fable-5-is-worth-the-premium", "published_at": "2026-06-14T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f00361e17c0", "title": "Opus 4.8 vs Opus 4.7 vs GPT-5.5 vs Composer 2.5 - 50 Real PRs in Go and Rust", "url": "https://www.stet.sh/blog/opus-48-vs-gpt-55-vs-opus-47-vs-composer-25", "published_at": "2026-06-02T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f0036e12603", "title": "I had Codex iterate on its own AGENTS.md 8 times and measured each version against real PRs. The best one still regressed on a clean holdout.", "url": "https://www.stet.sh/blog/how-i-used-codex-to-improve-its-own-agents-md", "published_at": "2026-05-27T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f00370b4aec", "title": "GPT-5.5 High Regression Check on GraphQL-go-tools", "url": "https://www.stet.sh/blog/gpt-55-high-regression-check-graphql-go-tools", "published_at": "2026-05-18T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f0037407451", "title": "Opus 4.7 Low Vs Medium Vs High Vs Xhigh Vs Max: the Reasoning Curve on 29 Real Tasks from an Open Source Repo", "url": "https://www.stet.sh/blog/opus-47-graphql-reasoning-curve", "published_at": "2026-05-12T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f003775cd6e", "title": "GPT-5.5 low vs medium vs high vs xhigh: the reasoning curve on 26 real tasks from an open source repo", "url": "https://www.stet.sh/blog/gpt-55-codex-graphql-reasoning-curve", "published_at": "2026-05-07T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f0037906dd6", "title": "GPT-5.5 vs GPT-5.4 vs Opus 4.7 on 56 real coding tasks from 2 open source repos", "url": "https://www.stet.sh/blog/gpt-55-vs-opus-47", "published_at": "2026-05-01T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f00381e466b", "title": "Opus 4.7 vs Old Opus 4.6 vs New Opus 4.6", "url": "https://www.stet.sh/blog/opus-4-7-zod", "published_at": "2026-04-17T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f0038543d72", "title": "Your AGENTS.md is the highest-leverage code you're not testing", "url": "https://www.stet.sh/blog/agents-md-testing", "published_at": "2026-04-08T12:00:00+00:00" }, { "id": "01a087e6-0359-7198-88f4-7f0038eb12d7", "title": "Your AI coding benchmark is hiding a 2x quality gap", "url": "https://www.stet.sh/blog/both-pass", "published_at": "2026-03-14T12:00:00+00:00" } ] posts Claim your blog

Back to stet.sh

Blog · corpus.blog/blogs/stet.sh/posts

stet.sh

stet.sh

2026

On 25 Stet Tasks, Opus 4.8 Wrote Smaller Patches. Opus 5 Searched Wider.original ↗

25 Aug 2026

I Tested Six Ways to Save $$$ on 5.6 Sol. The only one that worked was using Terra original ↗

20 Jul 2026

Sonnet 5 vs Opus 4.8: how they behave and when to use each original ↗

8 Jul 2026

GLM 5.2 on 50 real Go and Rust PRs: last on quality, and not the cheapest original ↗

19 Jun 2026

Composer 2.5 was 6.5x cheaper on 50 real Rust and Go PRs. Here's where it's actually safe.original ↗

18 Jun 2026

When Fable 5 Is Worth the Premium original ↗

14 Jun 2026

Opus 4.8 vs Opus 4.7 vs GPT-5.5 vs Composer 2.5 - 50 Real PRs in Go and Rust original ↗

2 Jun 2026

I had Codex iterate on its own AGENTS.md 8 times and measured each version against real PRs. The best one still regressed on a clean holdout.original ↗

27 May 2026

GPT-5.5 High Regression Check on GraphQL-go-tools original ↗

18 May 2026

Opus 4.7 Low Vs Medium Vs High Vs Xhigh Vs Max: the Reasoning Curve on 29 Real Tasks from an Open Source Repo original ↗

12 May 2026

GPT-5.5 low vs medium vs high vs xhigh: the reasoning curve on 26 real tasks from an open source repo original ↗

7 May 2026

GPT-5.5 vs GPT-5.4 vs Opus 4.7 on 56 real coding tasks from 2 open source repos original ↗

1 May 2026

Opus 4.7 vs Old Opus 4.6 vs New Opus 4.6 original ↗

17 Apr 2026

Your AGENTS.md is the highest-leverage code you're not testing original ↗

8 Apr 2026

Your AI coding benchmark is hiding a 2x quality gap original ↗

14 Mar 2026