Claude Haiku 5.5 vs Gemini 4 Carbon

vs

什麼情況選哪一個

claude-haiku-5-5 和 gemini-4-carbon 都能處理文字與圖片,且上下文範圍約為一百萬 token (分別為 1000000 與 1048576),但兩者的定價差異極大:前者輸入 $0.1、輸出 $0.5,相較於後者的 $2 與 $10,兩項價格皆相差 20 倍,且快取讀取費率為 $0.01 對比 $0.1。針對需要大量處理文字與圖片、回覆長度高達 128000 token,或是需要關閉思考功能的任務,請選擇 claude-haiku-5-5。若輸入包含 claude-haiku-5-5 不支援的音訊或影片,請選擇 gemini-4-carbon。

Benchmark 成績

Claude Haiku 5.5:供應商沒有公布 benchmark 成績。

高於平均沒有模型更高Gemini 4 Carbon12 / 1210 / 12
Claude Haiku 5.5 Gemini 4 Carbon 其他有成績的模型 其他模型平均 ★ 沒有模型分數更高
FrontierSWE 2 Proximal leaderboard, Carbon report
N/A
61.2%
OSWorld 2.1 82 offline tasks, partial score. Gemini: best of 3. Peers: published evaluations
N/A
81.5%
AutomationBench (Zapier) Zapier private set
N/A
沒有模型分數更高 53.9%
GraphWalks BFS (F1) 256k to 1M tokens, 200 tasks, Google evaluation
N/A
沒有模型分數更高 96.2%
Blueprint-Bench 2 Similarity score, Andon Labs leaderboard
N/A
沒有模型分數更高 58.6%
Vending-Bench 2 (USD) Andon Labs leaderboard, Carbon report · 5422-16484
N/A
沒有模型分數更高 16484
Chartography No tools, Carbon report
N/A
沒有模型分數更高 72.5%

供應商公布: Google

定價

Claude Haiku 5.5 Gemini 4 Carbon Δ
輸入 / 1M tokens $0.1 $2 0.05×
輸出 / 1M tokens $0.5 $10 0.05×
快取讀取 / 1M tokens $0.01 $0.1 0.1×
快取寫入 1.25x (5m) / 2x (1h) - -

費率取自網站建置時的即時目錄;各模型頁面都列有最新的價目。 主表顯示基礎級距。提示詞長度可能改變整筆請求的費率,各模型頁有完整價目。

Claude Haiku 5.5 · 依輸入長度分級的價格

整次請求依輸入長度所在級距計費。價格單位為美元/百萬 token。

輸入 token 數輸入輸出
0 - 100,000 $0.1 $0.5
100,001 - 1,000,000 $0.5 $2.5

兩者的相對位置:每 1M tokens 的輸入價格,涵蓋同一計費單位下全部 86 個聊天模型(對數尺度)

功能

Claude Haiku 5.5 Gemini 4 Carbon
工具使用 是 是
思考控制 可設定 一律開啟
結構化輸出 是 是
提示詞快取 顯式(由你標記前綴) 隱式(自動)
快取存活時間 5m default, 1h option 未公布
最小快取前綴 512 個 token 4,096 個 token

規格

Claude Haiku 5.5 Gemini 4 Carbon
輸入模態 文字 圖像 文字 圖像 音訊 影片
輸出模態 文字 文字
發布日期 2026-10-07 2026-11-10
知識截止日期 2026-06 2026-03
上下文視窗 1,000,000 1,048,576
最大輸出 128,000 65,536
思考參數 output_config.effort thinking_level
可接受的值
output_config.effort
  • low
  • medium
  • high
  • xhigh
  • max
thinking_level
  • low
  • medium
  • high

"minimal" and thinking_budget return 400

預設值 medium medium

規格摘錄自各供應商的文件;供應商沒有公布的項目就直接略過,不自行推測。 完整來源: Claude Haiku 5.5 · Gemini 4 Carbon

同一段提示詞,兩個模型,經閘道實測

提示詞 Explain to a CFO, in exactly three sentences, why our LLM bill doubled after we enabled reasoning. 檢查 恰好 3 句

Claude Haiku 5.5 通過 · 3 sentences

Enabling reasoning makes the model generate a long internal chain of thought before its answer, and those thinking tokens are billed as output tokens, which cost several times more than input tokens. The model often produces many times more output per request than it did without reasoning, so the per-call cost rises even when the visible answer looks about the same length. Because that extra spend scales with every request, the bill roughly doubled, and we can control it by setting reasoning budgets, using reasoning only for tasks that need it, or routing simpler requests to a non-reasoning model.

輸出 332 tok (+159 思考) 延遲 2.9 s

Gemini 4 Carbon 通過 · 3 sentences

Enabling reasoning instructs the AI to generate a lengthy, behind-the-scenes "thought process" to verify its logic before delivering a final answer. Because our vendor bills us based on total output volume, we are charged for all of that invisible scratchpad work in addition to the visible response, effectively doubling the compute consumed per query. Moving forward, we can control this cost by restricting reasoning strictly to complex, high-stakes workflows where higher accuracy justifies the premium, while routing routine tasks back to standard, lower-cost models.

輸出 908 tok (+799 思考) 延遲 18.0 s

是否照指令做(剛好三句,數得出來)、語氣是否貼合對象(寫給 CFO 看),以及下方 token 計量顯示出的隱藏思考計費落差。

提示詞 This function is supposed to return the longest run of consecutive equal items, but callers report it is off by one on single-element inputs and misses runs that end at the last index. Fix it and explain the bug in one paragraph. 檢查 修正後通過測試

Claude Haiku 5.5 通過 · 8/8 cases

```python def longest_run(xs): if not xs: return 0 best = cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: cur = 1 best = max(best, cur) return best ``` The bug is that `cur` counts adjacent equal *pairs*, not run *elements*. A run of k equal items produces k−1 matching pairs, so the function returns one less than the true length, and a single-element input produces no pairs at all, so it returns 0 instead of 1. Resetting `cur` to 0 on a mismatch compounds this by starting every run one step behind. The fix starts both `cur` and `best` at 1 (a non-empty list always has a run of at least one element), resets `cur` to 1 on a mismatch, and updates `best` after every element rather than only inside the equal-branch. The empty-list guard keeps the function from returning 1 for `[]`. […]

輸出 1012 tok (+591 思考) 延遲 6.0 s

Gemini 4 Carbon 通過 · 8/8 cases

```python def longest_run(xs): if not xs: return 0 best = 1 cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 best = max(best, cur) else: cur = 1 return best ``` The original function counts the number of matching adjacent *transitions* rather than the number of *elements* in a run because `best` and `cur` are initialized and reset to `0` instead of `1`. Because a single item on its own already forms a run of length `1`, starting at `0` causes every non-empty result to be short by one: single-element lists (where the loop never executes) and lists with no duplicates return `0` instead of `1`, and any run of $k$ equal items—including one ending at the last index—only increments `cur` $k - 1$ times, making it appear as though the final element of the run was missed. […]

輸出 1411 tok (+1131 思考) 延遲 32.7 s

修正是否真的正確(跑得起來)、說明的資訊密度,以及在範圍明確的任務上的 token 效率。

提示詞 Extract a JSON object with fields {company, quarter, revenue_usd_m, growth_pct, guidance} from this text. Use null for anything not stated; add no other fields. 檢查 有效的 JSON,完全符合 schema

Claude Haiku 5.5 通過 · 5/5 fields, guidance "Formal guidance withheld pending the Q4 "

```json { "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": null, "growth_pct": 11, "guidance": "Formal guidance withheld pending the Q4 close" } ``` Notes: The text gives revenue as 412 but doesn't state the currency, so I left `revenue_usd_m` as null. Growth is "eleven and change," so 11 is an approximation.

輸出 725 tok (+572 思考) 延遲 4.6 s

Gemini 4 Carbon 通過 · 5/5 fields, guidance null

```json { "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": null, "growth_pct": null, "guidance": null } ```

輸出 3338 tok (+3281 思考) 延遲 61.6 s

是否遵守 schema(不自創欄位)、面對幻覺誘因的表現(題目明確不給 guidance),以及結構化輸出路徑的差異。

提示詞 Write a 120-word product announcement for an API gateway feature that caches prompts across providers. Forbidden words: "seamless", "unlock", "game-changer", "revolutionize", "empower". Exactly one sentence must be a question. 檢查 120 詞,0 個禁用詞

Claude Haiku 5.5 通過 · 120 words, 0 banned, 1 question

**Introducing Prompt Cache Sync, the newest feature in GatewayOne.** Teams routing requests across OpenAI, Anthropic, and Google have often paid to process the same long system prompts again and again. Now the gateway stores those prompts once and reuses them wherever they appear, regardless of provider. Repeated context is matched by content hash, so cached prefixes return faster and cost less, with no changes to your application code. How much could this save your team? Enable Prompt Cache Sync from the dashboard, and your next billing report will show the difference. Cache entries follow your existing retention settings and can be purged per project at any time. Availability begins this week for all Scale and Enterprise plans worldwide. Start today.

輸出 2164 tok (+1929 思考) 延遲 9.9 s

Gemini 4 Carbon 通過 · 120 words, 0 banned, 1 question

Are your engineering teams paying multiple AI vendors to generate identical responses? Today, we are introducing Universal Prompt Caching for our flagship API Gateway. When applications route traffic across providers like OpenAI, Anthropic, and Mistral, duplicate queries drain budgets and degrade user experiences. Our new unified caching layer intercepts incoming prompts directly at the network edge. If an identical request was previously fulfilled by any connected model, the gateway delivers the stored response in milliseconds, completely bypassing external API calls. This practical upgrade cuts monthly token expenses substantially while accelerating overall application speed. Best of all, deployment requires zero complex code changes. Simply activate this capability inside your gateway configuration dashboard today to start reducing latency and infrastructure costs immediately.

輸出 2620 tok (+2480 思考) 延遲 31.1 s

是否遵守限制(字數上限、禁用詞清單、只能有一句問句)、文字風格的特徵,以及長度控制。

改一行程式碼就能在兩者之間切換

下面每個頁籤都列了這兩個模型 ID,要改的只有醒目標示的那兩行。端點、金鑰和請求格式都不變。

from openai import OpenAI

client = OpenAI(
    base_url="https://synthorai.io/v1",
    api_key="sk-syn-...",
)

resp = client.chat.completions.create(
    model="claude-haiku-5-5",
    # model="gemini-4-carbon",  # 取消這一行的註解,並把上一行註解掉
    messages=[{"role": "user", "content": "Summarize this diff"}],
    reasoning_effort="medium",
)
print(resp.choices[0].message.content)

取得 API 金鑰 →

常見問題

Claude Haiku 5.5 和 Gemini 4 Carbon 哪個比較便宜?

主表顯示基礎級距。提示詞長度可能改變整筆請求的費率,各模型頁有完整價目。

可以只串接一次,就對 Claude Haiku 5.5 和 Gemini 4 Carbon 做 A/B 測試嗎?

可以。兩個模型都走同一個 OpenAI 相容端點,用的也是同一把 API 金鑰,切換時只要改一行裡的模型名稱字串。你可以把一部分流量分別導到兩邊,再直接比較帳單。

Claude Haiku 5.5 與 Gemini 4 Carbon 支援提示詞快取嗎?

支援。兩者的快取讀取費率都低於輸入費率,所以前綴已經進快取的工作負載,實際成本會比官網價算出來的低。確切的快取讀取價格請見上方定價表。

相關比較