diff --git a/README.ja.md b/README.ja.md index fe19550b8..52557b0df 100644 --- a/README.ja.md +++ b/README.ja.md @@ -85,6 +85,24 @@ Claude Code、Codex、数々のOSSモデルに振り回されていませんか OmOをインストールして、`ultrawork`とタイプしてください。それだけです。 +## 私たちが賭ける未来 + +AIの未来について、明確な信念を持っています。 + +**モデルはどんどん安くなっている。** 一年前の最上位モデルが今日では無料で使える時代です。推論コストの急激な低下は、もはや不可逆なトレンドです。 + +**モデルはどんどん賢くなっている。** GPT-4レベルの性能が当たり前になり、次々と新たな能力が解き放たれています。賢さの向上も、同様に不可逆です。 + +**特定のプロバイダーが独占しない未来が来る。** OpenAI、Anthropic、Google、中国のモデル、オープンソースのモデル。それぞれが独自の強みを持ち、どれもが不可欠なツールとなっています。 + +**私たちはすべてのモデルを活用する。** 単一のプロバイダーに依存するのではなく、タスクに最適なモデルを選び、組み合わせ、オーケストレーションします。これが、真のパフォーマンスを引き出す唯一の方法です。 + +**モデルが専門化すればするほど、このアーキテクチャの価値は高まる。** コーディングに特化したモデル、推論に長けたモデル、創造性に優れたモデル。それぞれを適材適所で使い分けることで、初めて「チームとしてのAI」が実現します。 + +**私たちはその未来に賭けている。** + +Oh My OpenCodeは、マルチモデル時代のためのインフラストラクチャです。プロバイダーの囲い込みから解放され、常に最適なツールを選択し続ける。それが私たちの約束です。 + ## インストール ### 人間向け diff --git a/README.ko.md b/README.ko.md index 6f263333e..c58ddff0b 100644 --- a/README.ko.md +++ b/README.ko.md @@ -85,6 +85,26 @@ Claude Code, Codex, 온갖 OSS 모델들 사이에서 헤매고 있나요. 워 OmO 설치하고. `ultrawork` 치세요. 끝. +## 우리가 보는 미래 + +모델은 계속 싸지고 있다. +모델은 계속 똑똑해지고 있다. + +Anthropic이 독재하던 시대는 끝났다. OpenAI도, Google도, 어떤 단일 프로바이더도 전체를 집어삼키지 못할 것이다. + +우리는 그 미래에 베팅하고 있다. + +모든 모델을 활용한다. Kimi가 프론트엔드에서 빛나면 쓰고, GPT-5.3 Codex가 깊은 리서치를 잘하면 쓰고, Claude가 논리가 날카로우면 쓴다. 프로바이더 충성도 같은 건 없다. 오직 결과만 있다. + +모델들이 전문화될수록 이 아키텍처의 가치는 커진다. 하나의 거대한 모델에게 모든 걸 맡기던 시대는 가고, 각자의 강점이 뚜렷한 전문가 모델들이 나타나고 있다. 우리는 그 흐름을 타고 있다. + +**우리는 베팅했다. 모든 모델을 아우르는 오케스트레이션이 곧 표준이 될 것이라고.** + +OmO는 그 미래를 지금 여기서 구현한다. + +--- + + ## 설치 ### 사람용 diff --git a/README.md b/README.md index a682c36a1..bf6b69b76 100644 --- a/README.md +++ b/README.md @@ -88,6 +88,25 @@ We did the work. Tested everything. Kept what actually shipped. Install OmO. Type `ultrawork`. Done. +## The Bet + +We're betting on a future where models are cheaper, smarter, and no single provider dominates. +That future is arriving faster than anyone expected. + +**Models get cheaper every month.** What cost $20 last year costs $2 now. The race to the bottom is real, and we're here for it. +**Models get smarter every month.** Today's cheap model beats last year's flagship. The gap between frontier and affordable is collapsing. +**No one wins.** Anthropic, OpenAI, Google, DeepSeek, Alibaba - they're all building incredible models. The future isn't locked to one provider. It's a multi-model world. + +We built for that future. + +Sisyphus doesn't care if he's running on Claude, Kimi, or GLM. He picks the right model for the task. Frontend work? Kimi K2.5. Deep reasoning? Claude Opus. Code generation? GPT-5.3 Codex. The architecture treats models as interchangeable commodities, not sacred cows. + +This gets more valuable as models specialize. When Gemini dominates visual tasks and Qwen crushes long-context, we plug them in. Zero rewrites. Zero lock-in. + +Anthropic wants you locked in. We're building for the open market. + +The harness that wins isn't the one married to a single model. It's the one that rides them all. + ## Installation ### For Humans diff --git a/README.zh-cn.md b/README.zh-cn.md index a61127024..1bbb92239 100644 --- a/README.zh-cn.md +++ b/README.zh-cn.md @@ -85,6 +85,25 @@ 安装 OmO。敲下 `ultrawork`。完事。 +--- + +## 我们押注的未来 + +模型正在变得越来越便宜。 +模型正在变得越来越聪明。 +没有一个提供商能够垄断未来。 + +我们利用所有模型。 + +随着模型专业化，这种架构的价值会越来越大。Claude 擅长某些事，Kimi 擅长另一些，GPT-5.3 Codex 在代码上独树一帜。未来不是押注一个赢家，而是让正确的模型做正确的事。 + +我们正在押注那个未来。 + +这不是一个框架。这是一场对 AI 基础设施的重新想象。 + +--- + + ## 安装 ### 给人类看的 diff --git a/docs/guide/overview.md b/docs/guide/overview.md index 1c4379750..d3d820574 100644 --- a/docs/guide/overview.md +++ b/docs/guide/overview.md @@ -42,19 +42,29 @@ Hephaestus runs on GPT-5.3 Codex. Give him a goal, not a recipe. He explores the Use Hephaestus when you need deep architectural reasoning, complex debugging across many files, or cross-domain knowledge synthesis. Switch to him explicitly when the work demands GPT-5.3 Codex's particular strengths. +**Why this beats vanilla Codex CLI:** + + **Multi-model orchestration.** Pure Codex is single-model. OmO routes different tasks to different models automatically. GPT for deep reasoning. Gemini for frontend. Haiku for speed. The right brain for the right job. + **Background agents.** Fire 5+ agents in parallel. Something Codex simply cannot do. While one agent writes code, another researches patterns, another checks documentation. Like a real dev team. + **Category system.** Tasks are routed by intent, not model name. `visual-engineering` gets Gemini. `ultrabrain` gets GPT-5.3 Codex. `quick` gets Haiku. No manual juggling. + **Accumulated wisdom.** Subagents learn from previous results. Conventions discovered in task 1 are passed to task 5. Mistakes made early aren't repeated. The system gets smarter as it works. + --- -## Better Than Pure Codex +## Better Than Pure Claude Code -Sisyphus + Hephaestus outperform vanilla Codex CLI. Here's why: +Claude Code is good. But it's a single agent running a single model doing everything alone. -**Multi-model orchestration.** Pure Codex is single-model. Oh My OpenCode routes different tasks to different models automatically. GPT for deep reasoning. Gemini for frontend. Haiku for speed tasks. The right brain for the right job. +Oh My OpenCode turns that into a coordinated team: -**Background agents.** Fire 5+ agents in parallel. Something Codex simply cannot do. While one agent writes code, another researches patterns, another checks documentation. Like a real dev team. + **Parallel execution.** Claude Code processes one thing at a time. OmO fires background agents in parallel — research, implementation, and verification happening simultaneously. Like having 5 engineers instead of 1. + **Hash-anchored edits.** Claude Code's edit tool fails when the model can't reproduce lines exactly. OmO's `LINE#ID` content hashing validates every edit before applying. Grok Code Fast 1 went from 6.7% to 68.3% success rate just from this change. + **Intent Gate.** Claude Code takes your prompt and runs. OmO classifies your true intent first — research, implementation, investigation, fix — then routes accordingly. Fewer misinterpretations, better results. + **LSP + AST tools.** Workspace-level rename, go-to-definition, find-references, pre-build diagnostics, AST-aware code rewrites. IDE precision that vanilla Claude Code doesn't have. + **Skills with embedded MCPs.** Each skill brings its own MCP servers, scoped to the task. Context window stays clean instead of bloating with every tool. + **Discipline enforcement.** Todo enforcer yanks idle agents back to work. Comment checker strips AI slop. Ralph Loop keeps going until 100% done. The system doesn't let the agent slack off. -**Category system.** Tasks are routed by intent, not model name. `category="visual-engineering"` gets Gemini. `category="ultrabrain"` gets GPT-5.3 Codex. `category="quick"` gets Haiku. No manual juggling. - -**Accumulated wisdom.** Subagents learn from previous results. Conventions discovered in task 1 are passed to task 5. Mistakes made early aren't repeated. The system gets smarter as it works. +**The fundamental advantage.** Models have different temperaments. Claude thinks deeply. GPT reasons architecturally. Gemini visualizes. Haiku moves fast. Single-model tools force you to pick one personality for all tasks. Oh My OpenCode leverages them all, routing by task type. This isn't a temporary hack — it's the only architecture that makes sense as models specialize further. The gap between multi-model orchestration and single-model limitation widens every month. We're betting on that future. --- @@ -64,7 +74,7 @@ Before acting on any request, Sisyphus classifies your true intent. Are you asking for research? Implementation? Investigation? A fix? The Intent Gate figures out what you actually want, not just the literal words you typed. This means the agent understands context, nuance, and the real goal behind your request. -Regular Codex doesn't have this. It takes your prompt and runs. Oh My OpenCode thinks first, then acts. +Claude Code doesn't have this. It takes your prompt and runs. Oh My OpenCode thinks first, then acts. ---