OverpayingForAIPricing desk

Summarise · 3 checks · max 400 tokens · benched 2026-09-16

GPT-5.5 vs GPT-5.4 vs GPT-5.4 mini vs GPT-5.4 Nano vs GPT-5.3-Codex on one-sentence changelog summary

OpenAI models side by side on "One-sentence changelog summary": GPT-5.4 Nano scores 9/10; GPT-5.4 Nano is the cheapest answer scoring 8+ at $0.09 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.

The prompt every model received

User

Summarise this release note in one sentence of at most 30 words. Output only the sentence.

Release 4.7.0: The public API now enforces a rate limit of 600 requests per minute per key (previously unlimited). Requests over the limit return HTTP 429 with a Retry-After header. Webhook deliveries are now signed with HMAC-SHA256 and the old unsigned mode will be removed in 4.9. We also fixed a bug where exporting more than 10,000 rows produced an empty file.

Rubric for the judge: A single accurate sentence that prioritises the rate limit and signed webhooks; no bullet list, no preamble.

Side by side

Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.

GPT-5.5

openai/gpt-5.5

9/10
Latency
483ms
Cost
$0.00202
Per 1,000
$2.02

117 in · 48 out (18 reasoning) · 15 words · checks 2/3

  • max_words
  • contains
  • regex
Release 4.7.0 adds API rate limits, signed webhooks, and fixes large exports returning empty files.

Judge: Accurate single sentence under 30 words (14 words) that prioritises rate limits and signed webhooks as required, with no preamble or bullet list.

GPT-5.4

openai/gpt-5.4

9/10
Latency
408ms
Cost
$0.00127
Per 1,000
$1.27

117 in · 65 out (11 reasoning) · 22 words · checks 2/3

  • max_words
  • contains
  • regex
Release 4.7.0 adds a 600-requests-per-minute API rate limit with 429/Retry-After responses, signs webhooks with HMAC-SHA256, and fixes empty exports over 10,000 rows.

Judge: Accurate single sentence under 30 words (27 words), prioritizes rate limit and signed webhooks as required, no preamble or bullets, though could have been slightly more explicit about the deprecation timeline.

GPT-5.4 mini

openai/gpt-5.4-mini

9/10
Latency
373ms
Cost
$0.00088
Per 1,000
$0.88

117 in · 177 out (131 reasoning) · 19 words · checks 1/3

  • max_words
  • contains
  • regex
Release 4.7.0 adds a 600-requests-per-minute public API limit, signs webhooks with HMAC-SHA256, and fixes empty exports over 10,000 rows.

Judge: Accurate single sentence under 30 words (19 words), prioritizes rate limit and signed webhooks as required, no preamble or bullets, though could have mentioned the HTTP 429 response for completeness.

GPT-5.4 Nano

openai/gpt-5.4-nano

9/10
top judge scorecheapest scoring 8+
Latency
283ms
Cost
$8.7e-5
Per 1,000
$0.09

117 in · 51 out · 23 words · checks 2/3

  • max_words
  • contains
  • regex
Release 4.7.0 adds a per-key 600-requests/min rate limit (HTTP 429 with Retry-After), signs webhook deliveries with HMAC-SHA256, and fixes exporting over 10,000 rows.

Judge: Accurate single sentence under 30 words (27 words), prioritizes rate limit and signed webhooks as required, no preamble or bullet list, minor: could emphasize deprecation of unsigned mode more explicitly.

GPT-5.3-Codex

openai/gpt-5.3-codex

9/10
Latency
408ms
Cost
$0.00383
Per 1,000
$3.83

117 in · 259 out (196 reasoning) · 29 words · checks 2/3

  • max_words
  • contains
  • regex
Release 4.7.0 adds a 600-requests/minute API key rate limit with HTTP 429/Retry-After, introduces HMAC-SHA256 webhook signing ahead of unsigned removal in 4.9, and fixes empty exports over 10,000 rows.

Judge: Accurate single sentence under 30 words (29 words), prioritizes rate limit and signed webhooks as required, no preamble or bullet list, minor: could emphasize the deprecation timeline slightly more clearly.

Frequently asked

What does this prompt test?

Summarise: A single accurate sentence that prioritises the rate limit and signed webhooks; no bullet list, no preamble. The deterministic checks are max_words, contains, regex.

Which model should I pick for this task?

If the judge's bar of 8/10 is good enough for you, GPT-5.4 Nano at $0.09 per 1,000 runs. If you need the top score, GPT-5.4 Nano at $0.09 per 1,000 runs.

If our calculators helped you cut down on hidden AI wallet leaks, thanks for using them. A tiny fraction of your savings is what keeps our pricing indexes updated daily.

Not sure which AI is cheapest for your use case? Find out in 30 seconds — no signup required.

AI cost intelligence

Stop overpaying for AI tools

Join the OverpayingForAI list for pricing updates, cheaper alternatives, and practical buying guidance.

Now tracking 50+ AI tools, models, platforms, subscriptions, coding tools, and automation products.

We use your email only for OverpayingForAI updates. Unsubscribe anytime.