GPT-3.5 Turbo vs GPT-4
GPT-4 leads GPT-3.5 Turbo on the Artificial Analysis Intelligence Index, 6.7 to 5.5, and GPT-3.5 Turbo costs 40.0x less per output token ($1.50 vs $60.00 per million). Updated October 22, 2026; 9 benchmarks have a result for both.
GPT-3.5 Turbo or GPT-4: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick GPT-3.5 Turbo for output tokens 40.0x cheaper ($1.50 vs $60.00 per million) and a larger context window (16K against 8K tokens).
- Shared results
- 9
- GPT-3.5 Turbo only
- 4
- GPT-4 only
- 36
- Like-for-like categories
- 6 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
GPT-4
GPT-4 leads, 6.7 against 5.5.
Same board, both measured
Reasoning
GPQA Diamond
No call
Only GPT-3.5 Turbo has a GPQA Diamond result, 30.6%.
Directional only
Long documents
Listed context window
GPT-3.5 Turbo
GPT-3.5 Turbo takes 16K tokens against 8K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-3.5 Turbo
GPT-3.5 Turbo $0.0013, GPT-4 $0.060 per turn.
List prices
At a glance
- Coding index
- GPT-3.5 Turbo not benchmarked, GPT-4 not benchmarked.
- Widest category
- Math, 34.2 points apart on average over 1 shared benchmark.
- Widest single gap
- GSM8K: GPT-3.5 Turbo 57.8% against GPT-4 92.0%, 34.2 points apart on a 0 to 100 scale.
- Output price
- GPT-3.5 Turbo $1.50 against GPT-4 $60.00 per million output tokens, 40.0x apart.
- Context window
- GPT-3.5 Turbo takes 16K tokens, GPT-4 8K.
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-3.5 Turbo | Bars | GPT-4 | Ties |
|---|---|---|---|---|
| Coding | 0 | 1 | ||
| Intelligence | 0 | 1 | ||
| Reasoning | 0 | 2 | ||
| Math | 0 | 1 | ||
| Knowledge | 0 | 2 | ||
| General | 0 | 2 | ||
| All shared boards | 0 | 9 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index01
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| SWE-bench Verified | Not measured | 22.4% | Only GPT-4 measured |
| SWE-bench Lite | Not measured | 39.7% | Only GPT-4 measured |
| SWE-bench (full test set) | Not measured | 12.5% | Only GPT-4 measured |
| AA Coding Index | 10.7 | 13.1 | GPT-4 |
| LMArena Text: coding | Not measured | 1209 | Only GPT-4 measured |
| LMArena Text: software and IT | Not measured | 1195 | Only GPT-4 measured |
| SciCode | 0.0% | Not measured | Only GPT-3.5 Turbo measured |
| Neither model has a result on the other 32 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| AA Intelligence Index | 5.5 | 6.7 | GPT-4 |
| Epoch Capabilities Index | Not measured | 125.7 | Only GPT-4 measured |
Agentic workNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| Neither model has a result on 41 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| GPQA Diamond | 30.6% | Not measured | Only GPT-3.5 Turbo measured |
| GPQA Diamond (Artificial Analysis run) | 29.7% | 34.9% | GPT-4 |
| BIG-Bench Hard | 61.6% | 75.1% | GPT-4 |
| LMArena Text: hard prompts | Not measured | 1200 | Only GPT-4 measured |
| LMArena Text: expert | Not measured | 1149 | Only GPT-4 measured |
| LMArena Text: hard prompts in English | Not measured | 1209 | Only GPT-4 measured |
| Neither model has a result on the other 16 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index01
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| GSM8K | 57.8% | 92.0% | GPT-4 |
| LMArena Text: math | Not measured | 1230 | Only GPT-4 measured |
| LMArena Text: mathematical work | Not measured | 1228 | Only GPT-4 measured |
| Neither model has a result on the other 10 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | 64.4% | Not measured | Only GPT-3.5 Turbo measured |
| MedQA (Vals AI) | 58.5% | Not measured | Only GPT-3.5 Turbo measured |
| MMLU Pro (Artificial Analysis run) | 46.2% | 56.2% | GPT-4 |
| MMLU | 68.5% | 86.4% | GPT-4 |
| AI2 Reasoning Challenge | Not measured | 96.3% | Only GPT-4 measured |
| LMArena Text: science | Not measured | 1177 | Only GPT-4 measured |
| LMArena Text: medicine and healthcare | Not measured | 1145 | Only GPT-4 measured |
| LMArena Text: legal and government | Not measured | 1191 | Only GPT-4 measured |
| LMArena Text: business and finance | Not measured | 1145 | Only GPT-4 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-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | GPT-3.5 Turbo | GPT-4 | Leader |
|---|---|---|---|
| IFBench | Not measured | 33.2% | Only GPT-4 measured |
| DTBench | 48.5% | 62.7% | GPT-4 |
| LMCA | 9.7% | 17.1% | GPT-4 |
| HellaSwag | Not measured | 95.3% | Only GPT-4 measured |
| WinoGrande | Not measured | 87.5% | Only GPT-4 measured |
| LMArena Text | Not measured | 1206 | Only GPT-4 measured |
| LMArena Text: instruction following | Not measured | 1201 | Only GPT-4 measured |
| LMArena Text: creative writing | Not measured | 1191 | Only GPT-4 measured |
| LMArena Text: English | Not measured | 1220 | Only GPT-4 measured |
| LMArena Text: non-English | Not measured | 1194 | Only GPT-4 measured |
| LMArena Text: Chinese | Not measured | 1184 | Only GPT-4 measured |
| LMArena Text: German | Not measured | 1198 | Only GPT-4 measured |
| LMArena Text: Japanese | Not measured | 1137 | Only GPT-4 measured |
| LMArena Text: Korean | Not measured | 1087 | Only GPT-4 measured |
| LMArena Text: French | Not measured | 1220 | Only GPT-4 measured |
| LMArena Text: Spanish | Not measured | 1197 | Only GPT-4 measured |
| LMArena Text: Russian | Not measured | 1195 | Only GPT-4 measured |
| LMArena Text: Polish | Not measured | 1265 | Only GPT-4 measured |
| LMArena Text: multi-turn | Not measured | 1206 | Only GPT-4 measured |
| LMArena Text: longer queries | Not measured | 1190 | Only GPT-4 measured |
| LMArena Text: writing and language | Not measured | 1219 | Only GPT-4 measured |
| LMArena Text: entertainment and media | Not measured | 1184 | Only GPT-4 measured |
| Neither model has a result on the other 24 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-3.5 Turbo | GPT-4 | 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-3.5 Turbo costs less
Repository review
50K input + 3K output tokens
GPT-3.5 Turbo costs less
Agent loop
200K input + 10K output tokens
GPT-3.5 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-3.5 Turbo | OpenAI GPT-4 |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | May 28, 2023 | May 28, 2023 |
| Context window | 16K | 8K |
| Input $/Mtok | $0.50 | $30.00 |
| Output $/Mtok | $1.50 | $60.00 |
| 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-3.5 Turbo better than GPT-4 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick GPT-3.5 Turbo for output tokens 40.0x cheaper ($1.50 vs $60.00 per million) and a larger context window (16K against 8K tokens).
Which is cheaper, GPT-3.5 Turbo or GPT-4?
GPT-3.5 Turbo costs $0.50 per million input tokens and $1.50 per million output tokens; GPT-4 costs $30.00 and $60.00. Cost per solved task is known for neither of them.
Which has the bigger context window, GPT-3.5 Turbo or GPT-4?
GPT-3.5 Turbo, with 16K tokens against 8K.
How many benchmarks is the GPT-3.5 Turbo vs GPT-4 comparison based on?
9 benchmarks have a result for both models as of October 22, 2026. GPT-3.5 Turbo's coding index rests on 0 independent benchmarks and GPT-4's on 0. Where only one model has a result, the page says so instead of naming a winner.