DeepSeek V3.1 vs GPT-4
DeepSeek V3.1 leads GPT-4 on the Artificial Analysis Intelligence Index, 13.7 to 6.7, and DeepSeek V3.1 costs 40.0x less per output token ($1.50 vs $60.00 per million). Updated October 21, 2026; 3 benchmarks have a result for both.
DeepSeek V3.1 or GPT-4: which to pick
Neither model has a coding score yet, so the two cannot be ranked on coding.
Pick DeepSeek V3.1 for output tokens 40.0x cheaper ($1.50 vs $60.00 per million) and a larger context window (164K against 8K tokens).
- Shared results
- 3
- DeepSeek V3.1 only
- 8
- GPT-4 only
- 42
- 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
DeepSeek V3.1
DeepSeek V3.1 leads, 13.7 against 6.7.
Same board, both measured
Web design
Design Arena: websites
No call
Only DeepSeek V3.1 has a Design Arena: websites result, 1123.
Directional only
Long documents
Listed context window
DeepSeek V3.1
DeepSeek V3.1 takes 164K tokens against 8K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V3.1
DeepSeek V3.1 $0.0013, GPT-4 $0.060 per turn.
List prices
At a glance
- Coding index
- DeepSeek V3.1 not benchmarked, GPT-4 not benchmarked.
- Widest category
- General, 13.6 points apart on average over 2 shared benchmarks.
- Widest single gap
- DTBench: DeepSeek V3.1 82.7% against GPT-4 62.7%, 20.0 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V3.1 $1.50 against GPT-4 $60.00 per million output tokens, 40.0x apart.
- Context window
- DeepSeek V3.1 takes 164K 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 | DeepSeek V3.1 | Bars | GPT-4 | Ties |
|---|---|---|---|---|
| Intelligence | 1 | 0 | ||
| General | 2 | 0 | ||
| All shared boards | 3 | 0 |
CodingDirectional onlySix of these boards feed the Vibe Coding Index6 boards, none shared
| Benchmark | DeepSeek V3.1 | 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 | Not measured | 13.1 | Only GPT-4 measured |
| LMArena Text: coding | Not measured | 1209 | Only GPT-4 measured |
| LMArena Text: software and IT | Not measured | 1195 | Only GPT-4 measured |
| Neither model has a result on the other 33 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| AA Intelligence Index | 13.7 | 6.7 | DeepSeek V3.1 |
| 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 | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| Neither model has a result on 41 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index6 boards, none shared
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| GPQA Diamond (Artificial Analysis run) | Not measured | 34.9% | Only GPT-4 measured |
| BIG-Bench Hard | Not measured | 75.1% | Only GPT-4 measured |
| 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 |
| Kagi LLM Benchmark | 53.2% | Not measured | Only DeepSeek V3.1 measured |
| Neither model has a result on the other 16 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| GSM8K | Not measured | 92.0% | Only GPT-4 measured |
| 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 | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| MMLU Pro (Artificial Analysis run) | Not measured | 56.2% | Only GPT-4 measured |
| MMLU | Not measured | 86.4% | Only GPT-4 measured |
| 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 11 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index7 boards, none shared
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| Design Arena: websites | 1123 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: UI components | 1085 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: game dev | 1084 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: data viz | 1095 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: 3D | 1077 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: all code categories | 1112 | Not measured | Only DeepSeek V3.1 measured |
| Design Arena: SVG | 967 | Not measured | Only DeepSeek V3.1 measured |
| Neither model has a result on the other 2 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V3.1 | GPT-4 | Leader |
|---|---|---|---|
| IFBench | Not measured | 33.2% | Only GPT-4 measured |
| DTBench | 82.7% | 62.7% | DeepSeek V3.1 |
| LMCA | 24.3% | 17.1% | DeepSeek V3.1 |
| 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 | DeepSeek V3.1 | 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
DeepSeek V3.1 costs less
Repository review
50K input + 3K output tokens
DeepSeek V3.1 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V3.1 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 | DeepSeek V3.1 | OpenAI GPT-4 |
|---|---|---|
| Maker | DeepSeek | OpenAI |
| Weights | Open weights | No public weights listed |
| Released | August 21, 2025 | May 28, 2023 |
| Context window | 164K | 8K |
| Input $/Mtok | $0.55 | $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 DeepSeek V3.1 better than GPT-4 for coding?
Neither model has a coding score yet, so the two cannot be ranked on coding. Pick DeepSeek V3.1 for output tokens 40.0x cheaper ($1.50 vs $60.00 per million) and a larger context window (164K against 8K tokens).
Which is cheaper, DeepSeek V3.1 or GPT-4?
DeepSeek V3.1 costs $0.55 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, DeepSeek V3.1 or GPT-4?
DeepSeek V3.1, with 164K tokens against 8K.
How many benchmarks is the DeepSeek V3.1 vs GPT-4 comparison based on?
3 benchmarks have a result for both models as of October 21, 2026. DeepSeek V3.1'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.