Migrating from gpt-5.4 to gpt-5.5

What changes when you move from gpt-5.4 to gpt-5.5: input price +100%, output price +100%.

From: gpt-5.4
Status: Active
Shutdown: none announced
Time left: —

To: gpt-5.5
Status: Active
Shutdown: none announced
Time left: —

Prices

USD per 1M tokens, standard tier

Type	gpt-5.4	gpt-5.5	Change
Input	$2.50first seen October 1, 2026 · source	$5.00first seen October 1, 2026 · source	+100%
Cached input	$0.25first seen October 1, 2026 · source	$0.50first seen October 1, 2026 · source	+100%
Output	$15.00first seen October 1, 2026 · source	$30.00first seen October 1, 2026 · source	+100%

At 1M input and 250K output tokens a month: $6.25 on gpt-5.4, $12.50 on gpt-5.5 (+100%).

Other tiers, prompt-size bands and modalities are on each model's page.

Limits

Tokens

Limit	gpt-5.4	gpt-5.5	Change
Context window	1,050,000 tokensin place when tracking began (October 7, 2026) · source	1,050,000 tokensin place when tracking began (October 7, 2026) · source	0%
Max output tokens	128,000 tokensin place when tracking began (October 7, 2026) · source	128,000 tokensin place when tracking began (October 7, 2026) · source	0%

Benchmarks

Same benchmark and source on both sides; settings named where they differ

Benchmark	gpt-5.4	gpt-5.5	Change
Arena TextArena, high effort. Other settings 1465 ±4 → 1477 ±4	1475 ±4	1481 ±4	+6 within range
Arena Text: CodingArena, high effort. Other settings 1512 ±6 → 1511 ±6	1520 ±6	1519 ±6	−1 within range
Arena WebDevArena, xhigh effort, codex harness (vs high effort, codex harness). Ranges don't overlap. Other settings 1463 ±19 → 1487 ±6 high effort, codex harness not rated → 1456 ±6 codex harness	1463 ±19	1513 ±6	+50
Arena VisionArena. Other settings 1285 ±6 → 1281 ±6 high effort	1279 ±7	1286 ±6	+7 within range
Arena SearchArena. Ranges don't overlap.	1196 ±6	1226 ±5	+30
Arena DocumentArena. Ranges don't overlap. Other settings not rated → 1480 ±7 high effort	1470 ±6	1483 ±6	+13
Arena AgentArena, xhigh effort (vs high effort). Other settings not rated → -0.001 ±0.012	0.016 ±0.011	0.033 ±0.013	+0.017 within range
FrontierMath Tier 4Epoch AI, xhigh effort.	49.0%	72.5%	+23.5 pts
FrontierMath Tiers 1-3Epoch AI, xhigh effort.	78.6%	85.3%	+6.7 pts
GPQA DiamondEpoch AI, low effort. Other settings 74.7% → 77.3% no reasoning	84.8%	90.7%	+5.9 pts
OTIS Mock AIME 2024-2025Epoch AI, low effort. Other settings 57.8% → 57.8% no reasoning	84.4%	84.4%	0 pts
SimpleQA VerifiedEpoch AI, xhigh effort.	45.1%	63.0%	+17.9 pts

A score from one source is never compared with another source's, and company-reported scores are compared only within the same company; settings named where they differ. Data: Arena (arena.ai), CC BY 4.0; Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Features and parameters

Compare models

Same on both

Structured outputs: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Function calling: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Prompt caching: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Batch API: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Image input: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Reasoning / thinking: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Fine-tuning: not supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source
Built-in web search: supported on both.gpt-5.4: in place when tracking began (October 7, 2026) · source · gpt-5.5: in place when tracking began (October 7, 2026) · source

Not published for gpt-5.4: PDF input, temperature parameter, top_p parameter, logprobs.

Not published for gpt-5.5: PDF input, temperature parameter, top_p parameter, logprobs.

Aliases

All aliases

No aliases point at either model.

On cloud platforms

All platform dates

Model	Platform	Platform ID	Shutdown	Time left
gpt-5.4	Amazon Bedrock	openai.gpt-5.4	not before Jun 1, 2027Active
gpt-5.4	Microsoft Foundryversion 2026-03-05	gpt-5.4	Sep 2, 2027Active	322 days
gpt-5.5	Amazon Bedrock	openai.gpt-5.5	not before Jun 1, 2027Active
gpt-5.5	Microsoft Foundryversion 2026-04-24	gpt-5.5	Oct 26, 2027Active	376 days

Read daily from each platform's own pages. The status above uses each company's own date. Time left counts to 00:00 UTC on the shutdown date, which is the evening before in ET.

Checklist

Change the model string from gpt-5.4 to gpt-5.5.
Re-check costs: gpt-5.5 costs more than gpt-5.4 (input +100%, output +100%).

Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: gpt-5.4, gpt-5.5.