GPT-4 Turbo vs o3 Pro
o3 Pro leads GPT-4 Turbo on the Artificial Analysis Intelligence Index, 21.9 to 7.0, and GPT-4 Turbo costs 2.7x less per output token ($30.00 vs $80.00 per million). Updated October 21, 2026; 4 benchmarks have a result for both.
GPT-4 Turbo or o3 Pro: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick GPT-4 Turbo for output tokens 2.7x cheaper ($30.00 vs $80.00 per million) and work measured by AA Intelligence Index or Epoch Capabilities Index, where it leads.
- Shared results
- 4
- GPT-4 Turbo only
- 38
- o3 Pro only
- 7
- Like-for-like categories
- 2 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
No call
Neither has a coding score yet.
One side not measured
Intelligence
AA Intelligence Index
o3 Pro
o3 Pro leads, 21.9 against 7.0.
Same board, both measured
Long documents
Listed context window
o3 Pro
o3 Pro takes 200K tokens against 128K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-4 Turbo
GPT-4 Turbo $0.025, o3 Pro $0.060 per turn.
List prices
At a glance
- Coding index
- GPT-4 Turbo not benchmarked, o3 Pro not benchmarked.
- Widest category
- General, 27.0 points apart on average over 2 shared benchmarks.
- Widest single gap
- LMCA: GPT-4 Turbo 9.8% against o3 Pro 38.5%, 28.7 points apart on a 0 to 100 scale.
- Output price
- GPT-4 Turbo $30.00 against o3 Pro $80.00 per million output tokens, 2.7x apart.
- Context window
- GPT-4 Turbo takes 128K tokens, o3 Pro 200K.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-4 Turbo | Bars | o3 Pro | Ties |
|---|---|---|---|---|
| Intelligence | 0 | 2 | ||
| General | 0 | 2 | ||
| All shared boards | 0 | 4 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index6 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 84.9% | Only o3 Pro measured |
| AA Coding Index | 21.5 | Not measured | Only GPT-4 Turbo measured |
| LiveCodeBench (Artificial Analysis run) | 29.1% | Not measured | Only GPT-4 Turbo measured |
| WeirdML | Not measured | 58.2% | Only o3 Pro measured |
| LMArena Text: coding | 1268 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: software and IT | 1267 | Not measured | Only GPT-4 Turbo measured |
| Neither model has a result on the other 32 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| AA Intelligence Index | 7.0 | 21.9 | o3 Pro |
| Epoch Capabilities Index | 127.3 | 147.3 | o3 Pro |
Agentic workNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| Neither model has a result on 39 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 84.5% | Only o3 Pro measured |
| Humanity's Last Exam (Artificial Analysis run) | 3.1% | Not measured | Only GPT-4 Turbo measured |
| ARC-AGI | Not measured | 59.3% | Only o3 Pro measured |
| ARC-AGI-2 | Not measured | 4.9% | Only o3 Pro measured |
| LMArena Text: hard prompts | 1251 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: expert | 1223 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: hard prompts in English | 1266 | Not measured | Only GPT-4 Turbo measured |
| Neither model has a result on the other 14 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| LMArena Text: math | 1272 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: mathematical work | 1268 | Not measured | Only GPT-4 Turbo measured |
| Neither model has a result on the other 11 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | 80.5% | Not measured | Only GPT-4 Turbo measured |
| MedQA (Vals AI) | 82.0% | Not measured | Only GPT-4 Turbo measured |
| MMLU Pro (Artificial Analysis run) | 69.4% | Not measured | Only GPT-4 Turbo measured |
| MMLU | 81.3% | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: science | 1256 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: medicine and healthcare | 1239 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: legal and government | 1265 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: business and finance | 1222 | Not measured | Only GPT-4 Turbo measured |
| Neither model has a result on the other 9 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| DTBench | 61.6% | 86.9% | o3 Pro |
| LMCA | 9.8% | 38.5% | o3 Pro |
| Fiction.LiveBench | Not measured | 97.2% | Only o3 Pro measured |
| Creative writing (Lech Mazur) | Not measured | 84.4% | Only o3 Pro measured |
| LMArena Text | 1272 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: instruction following | 1249 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: creative writing | 1268 | Not measured | Only GPT-4 Turbo measured |
| LMArena Vision | 1090 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: English | 1291 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: non-English | 1245 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Chinese | 1242 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: German | 1259 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Japanese | 1194 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Korean | 1187 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: French | 1276 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Spanish | 1260 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Russian | 1259 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: Polish | 1302 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: multi-turn | 1267 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: longer queries | 1254 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: writing and language | 1279 | Not measured | Only GPT-4 Turbo measured |
| LMArena Text: entertainment and media | 1258 | Not measured | Only GPT-4 Turbo measured |
| LMArena Vision: English | 1097 | Not measured | Only GPT-4 Turbo measured |
| LMArena Vision: Chinese | 1045 | Not measured | Only GPT-4 Turbo measured |
| Neither model has a result on the other 22 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-4 Turbo | o3 Pro | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-4 Turbo costs less
Repository review
50K input + 3K output tokens
GPT-4 Turbo costs less
Agent loop
200K input + 10K output tokens
GPT-4 Turbo costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-4 Turbo | OpenAI o3 Pro |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | April 9, 2024 | June 10, 2025 |
| Context window | 128K | 200K |
| Input $/Mtok | $10.00 | $20.00 |
| Output $/Mtok | $30.00 | $80.00 |
| Coding index | Not benchmarked | Not benchmarked |
| Independent measurements behind it | 0 | 0 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-4 Turbo better than o3 Pro for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GPT-4 Turbo for output tokens 2.7x cheaper ($30.00 vs $80.00 per million) and work measured by AA Intelligence Index or Epoch Capabilities Index, where it leads.
Which is cheaper, GPT-4 Turbo or o3 Pro?
GPT-4 Turbo costs $10.00 per million input tokens and $30.00 per million output tokens; o3 Pro costs $20.00 and $80.00. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-4 Turbo or o3 Pro?
o3 Pro, with 200K tokens against 128K.
How many benchmarks is the GPT-4 Turbo vs o3 Pro comparison based on?
4 benchmarks have a result for both models as of October 21, 2026. GPT-4 Turbo's coding index rests on 0 independent benchmarks and o3 Pro's on 0. Where only one model has a result, the page says so instead of naming a winner.