早鳥價(12 月 25 日截止),只要持續訂閱就一直鎖定:Go 每月 $10(原價 $20),Plus 用 Opus 5.5 每月只要 $30(原價 $50)立即訂閱 →
文字對話1M 上下文

GPT-6.1 Sol

以 OpenAI 30% 的價格使用 GPT-6.1 Sol:每 100 萬輸入 $0.60、每 100 萬輸出 $3.00。比較 GPT-6 Sol 與 Astra,並提供 curl、Python 及 Codex 設定教學。

價格 / 100 萬 token
AIReiter
≤ 272,000 · 輸入$0.60
≤ 272,000 · 輸出$3.00
> 272,000 · 輸入$1.20
> 272,000 · 輸出$4.50
README
輸入
限制回覆長度。推理模型也會計入思考內容。
gpt-6.1-sol
輸出

A codebase you don't know is a risk, not a hurdle. The plan should burn down that risk in order: understand, run, trace, spike, then make the change small enough that "production-ready" is a property you can demonstrate, not a hope.

Here are the first five steps, with the evidence you collect at each.

Step 1 — Get it building and the baseline green (half day)
Read the README, the manifest(s), the CI config, and any architecture docs. Boot the app locally, run the full test suite, run the linter/typechecker.
Evidence: The exact commands that reproduce a clean build and a passing test suite from a fresh checkout (recorded output, not memory of it). A one-page map you wrote yourself: where the entry points are, the layers, the build/test toolchain, and the main risks to your feature (e.g. "this touches cron jobs and payments").

Step 2 — Trace the smallest end-to-end slice that resembles your feature (half day)
Don't start with your feature. Pick a neighboring one that already works and follow it from entry to persistence. Note the conventions along the way: how errors surface, how config is injected, how logging is done, how tests are written.
Evidence: An annotated trace — file → function → what it does — for that slice. A written list of "conventions I must follow" (not intuitions; things you observed in real code). A named shortlist of the 3–5 files you will actually touch. If you can't produce this trace, you're not ready to write code.

Step 3 — Spike the core path (day 2)
Build the ugly version: hardcoded values, no error handling, no tests. The purpose is to confirm the path you traced in step 2 is real and to surface what you didn't know you didn't know.
Evidence: A working spike demonstrating the feature's central data path, alongside a list of every assumption the spike broke and what you corrected. That correction list is the most valuable document in this whole plan.

Step 4 — Write the contract before the code (half of day 2 / day 3)
Once the spike proves the path, pin down what production needs: the inputs/outputs, the error cases, where it sits in the conventions from step 2. Then write the tests — they'll be red, but they're the specification.
Evidence: A one-to-two-page design doc, an agreed interface/API shape (with the team if there is one), and a red test suite that encodes intended behavior. If you can't write the contract without consulting the code, you haven't finished step 2.

Step 5 — Implement in small, verified increments (days 3–4)
Replace the spike with the real thing in small commits, each one keeping the suite green, leaning on the existing patterns. Run lint/tests/typecheck per commit — CI, not just locally — and exercise the actual path against a real instance (staging or a local environment that isn't stubbed).
Evidence: A branch with progressive commits, each green in CI; coverage on the new code; something that proves it works against reality (a test result, a log trace, a screenshot); and a review by at least one person who knows the codebase. The review counts as evidence — an unfamiliar codebase has tribal knowledge you cannot extract from the files alone.

Steps 6+ would be the things that actually make it "shipped": a migration plan and its rollback, feature flagging, observability, the release and post-release verification. But the first five get you to a reviewed, green, working slice in staging — which is the point at which you can say "this will work in production" with evidence behind it, instead of a guess.

生成時間42.7s
輸入 Token134
輸出 Token2,354
Token / 秒55.13
TTFT-

執行會以這段提示開啟 AI Chat。需要登入與點數。

模型詳情

在 Playground、API 請求與內部工作流程中使用相同的模型鍵。

模型 ID
gpt-6.1-sol
供應商
OpenAI
協定
OpenAI Chat Completions
上下文視窗
1,050,000 Token
最大輸出
128,000 Token

GPT-6.1 Sol 是什麼

GPT-6.1 Sol 比較

GPT-6.1 Sol 程式碼範例

01GPT-6.1 Sol 費用範例

用 3 個步驟試用 GPT-6.1 Sol

  1. 01

    設定推理強度

    建議從 medium 開始。針對回答前需要規劃的問題,可調高至 high、xhigh 或 max;當延遲比深度更重要時,則調低至 low。

  2. 02

    發送提示詞

    貼上您的實際任務而非玩具任務。每次回應下方都會回報 Token 使用量與消耗額度,讓您在正式採用前評估工作負載成本。

  3. 03

    複製 API 調用代碼

    使用模型 ID gpt-6.1-sol 將相同的請求帶入您的程式碼中。Tool calling 同時支援 Chat Completions 與 Responses API。

常見問題

01GPT-6.1 Sol 在 AIReiter 上的費用是多少?

GPT-6.1 Sol 在 AIReiter 上的費用為每 100 萬輸入 tokens $0.60、每 100 萬輸出 tokens $3.00,僅為 OpenAI 官方價格($2.00 與 $10.00)的 30%。快取輸入讀取為每 100 萬 $0.03(官方為 $0.10),快取寫入為每 100 萬 $0.75。超過 272K 輸入 tokens 的請求將按 2 倍輸入與 1.5 倍輸出費率計費。

02與 GPT-6 Sol 相比有哪些變化?

基礎輸入與輸出價格與 GPT-6 Sol 相同,但快取輸入價格減半,降至未快取費率的 5%(原為 10%)。Reasoning effort 新增了 max 設定,且不再支援 none 與 minimal 設定。Context window、輸出上限及長上下文附加費用則維持不變。

03GPT-6.1 Sol 比 GPT-6 Astra 更好嗎?

並非如此。Astra 仍是 OpenAI 最強大的 GPT-6 模型,而 GPT-6.1 Sol 則位居其下一個級別。選擇 Sol 的優勢在於成本:其短上下文輸入與輸出價格僅為 Astra 的五分之一(每 100 萬 tokens 輸入為 $2 對比 $10,輸出為 $10 對比 $50)。建議僅在 Sol 明顯無法滿足需求時才路由至 Astra。

04上下文視窗大小為何?

GPT-6.1 Sol 擁有 1,050,000 tokens 的 context window,最高支援 128,000 個輸出 tokens,輸入與輸出共享該額度。實際上,最划算的區間為前 272K 輸入 tokens:超過此範圍後,整個請求將按 2 倍輸入與快取費率,以及 1.5 倍輸出費率計費。

05支援哪些 reasoning effort 設定?

GPT-6.1 Sol 支援 low、medium、high、xhigh 與 max,預設為 medium。部分早期 GPT 模型支援的 none 與 minimal 設定已不再支援。較高的設定會產生更多計入輸出費用的 reasoning tokens,因此建議從 medium 開始,僅在需要複雜規劃的問題上調高。

06GPT-6.1 Sol 是否支援 tool calling?

是的。在 OpenAI 官方 API 上,GPT-6.1 Sol 僅能透過 Responses API 支援 tool calling,Chat Completions 則不支援。在 AIReiter 上,使用模型 gpt-6.1-sol 於 https://aireiter.com/api/v1/chat/completions 和 https://aireiter.com/api/v1/responses 皆可正常使用 tool calling,因此現有的 Chat Completions 工具呼叫流程可直接沿用而無需修改。

07如何在 Codex CLI 中使用 GPT-6.1 Sol?

在 ~/.codex/config.toml 中新增 aireiter 提供者,將 base_url 設為 https://aireiter.com/api/v1、wire_api 設為 responses、env_key 設為 AIREITER_API_KEY,並將 model 設為 gpt-6.1-sol。在該環境變數中 export 您的 AIReiter 金鑰並啟動 Codex。完整設定請參見上方的程式碼區塊。

08GPT-6.1 Sol 何時發布?

OpenAI 於 2026 年 9 月 29 日發布 GPT-6.1 Sol,作為 GPT-6 Sol 的升級版。它已在 OpenAI API 以及 ChatGPT Work 和 Codex 付費方案中提供。在 AIReiter 上,自推出起即可於 playground 和 API 中使用,價格僅為官方每 token 定價的 30%。