Migrating from gpt-3.5-turbo-0125 to gpt-4o-mini
What changes when you move from gpt-3.5-turbo-0125 to gpt-4o-mini: input price −70%, output price −60%, context window 16,385 → 128,000 tokens.
- Status
- Deprecated
- Shutdown
- Oct 23, 2026
- Time left
- 8 days
- Status
- Active
- Shutdown
- none announced
- Time left
- —
Prices
USD per 1M tokens, standard tier| Type | gpt-3.5-turbo-0125 | gpt-4o-mini | Change |
|---|---|---|---|
| Input | $0.50first seen October 1, 2026 · source | $0.15first seen October 1, 2026 · source | −70% |
| Cached input | Not published | $0.075first seen October 1, 2026 · source | — |
| Output | $1.50first seen October 1, 2026 · source | $0.60first seen October 1, 2026 · source | −60% |
At 1M input and 250K output tokens a month: $0.875 on gpt-3.5-turbo-0125, $0.30 on gpt-4o-mini (−65.7%).
Other tiers, prompt-size bands and modalities are on each model's page.
Limits
Tokens| Limit | gpt-3.5-turbo-0125 | gpt-4o-mini | Change |
|---|---|---|---|
| Context window | 16,385 tokensin place when tracking began (October 6, 2026) · source | 128,000 tokensin place when tracking began (October 6, 2026) · source | +681.2% |
| Max output tokens | 4,096 tokensin place when tracking began (October 6, 2026) · source | 16,384 tokensin place when tracking began (October 6, 2026) · source | +300% |
Benchmarks
Same benchmark, source and setting on both sides| Benchmark | gpt-3.5-turbo-0125 | gpt-4o-mini | Change |
|---|---|---|---|
| Arena TextArena. Ranges don't overlap. | 1226 ±5 | 1318 ±4 | +92 |
| Arena Text: CodingArena. Ranges don't overlap. | 1275 ±8 | 1348 ±7 | +73 |
| FrontierMath Tiers 1-3Epoch AI. | 0.0% | 0.7% | +0.7 pts |
| GPQA DiamondEpoch AI. | 27.2% | 37.7% | +10.5 pts |
| MATH Level 5Epoch AI. | 11.6% | 52.6% | +41.0 pts |
| OTIS Mock AIME 2024-2025Epoch AI. | 2.2% | 6.9% | +4.7 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 modelsSame on both
- Batch API: supported on both.gpt-3.5-turbo-0125: in place when tracking began (October 6, 2026) · source · gpt-4o-mini: in place when tracking began (October 6, 2026) · source
- Fine-tuning: supported on both.gpt-3.5-turbo-0125: in place when tracking began (October 6, 2026) · source · gpt-4o-mini: in place when tracking began (October 6, 2026) · source
Not published for gpt-3.5-turbo-0125: Structured outputs, Function calling, Prompt caching, Image input, PDF input, Reasoning / thinking, Built-in web search, temperature parameter, top_p parameter, logprobs.
Not published for gpt-4o-mini: Prompt caching, PDF input, Reasoning / thinking, temperature parameter, top_p parameter, logprobs.
Aliases
All aliasesAliases pointing at gpt-3.5-turbo-0125: gpt-3.5-turbo (in place when tracking began (October 7, 2026) · source)
No aliases point at gpt-4o-mini.
On cloud platforms
All platform dates| Model | Platform | Platform ID | Shutdown | Time left |
|---|---|---|---|---|
| gpt-4o-mini | Microsoft Foundryversion 2024-07-18 | gpt-4o-mini | Apr 14, 2027Deprecated | 181 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-3.5-turbo-0125togpt-4o-mini.
Counting to 8:00 pm ET on October 22, 2026, which is 00:00 UTC on October 23, 2026.
Compared from AIWire's published data when this page loaded. Each fact links to the page it came from. Model pages: gpt-3.5-turbo-0125, gpt-4o-mini.