DeepSeek V4 Pro 0423 vs gpt-oss-120b
Updated October 21, 2026. 55 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
DeepSeek V4 Pro 0423 and gpt-oss-120b are level for coding, 35.3 against 34.5 on the Vibe Coding Index, so price and context decide.
Pick gpt-oss-120b when output tokens cost 3.2x less, or when a solved task costs $0.51 against $0.57. Pick DeepSeek V4 Pro 0423 when the context window is larger, 1.05M against 131K tokens, or when what matters is AA Coding Index or LiveCodeBench (Vals AI), where it leads.
- Shared results
- 55
- DeepSeek V4 Pro 0423 only
- 28
- gpt-oss-120b only
- 23
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
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, four public coding benchmarks
No call
Level: 35.3 against 34.5, inside 2 points.
1 and 2 benchmarks behind the scores
Intelligence
AA Intelligence Index
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 30.4 against 11.6.
Same board, both measured
Agentic work
AA Agentic Index
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 26.3 against 3.7.
Same board, both measured
Reasoning
GPQA Diamond
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 89.4% against 78.5%.
Same board, both measured
Math
AIME
No call
Only gpt-oss-120b has a AIME result, 92.6%.
Directional only
Knowledge
MMLU Pro
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 87.3% against 79.2%.
Same board, both measured
Web design
Design Arena: websites
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 leads, 1235 against 968.
Same board, both measured
Long documents
Listed context window
DeepSeek V4 Pro 0423
DeepSeek V4 Pro 0423 takes 1.05M tokens against 131K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
gpt-oss-120b
DeepSeek V4 Pro 0423 $0.0019, gpt-oss-120b $0.0004 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
gpt-oss-120b
DeepSeek V4 Pro 0423 $0.57, gpt-oss-120b $0.51.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V4 Pro 0423 35.3 from 1 benchmark, gpt-oss-120b 34.5 from 2 benchmarks.
- Widest category
- Web design, 34.3 points apart on average over 6 shared benchmarks.
- Widest single gap
- Design Arena: 3D: DeepSeek V4 Pro 0423 1254 against gpt-oss-120b 901, 44.1 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V4 Pro 0423 $1.9105 against gpt-oss-120b $0.60 per million output tokens, 3.2x apart.
- Context window
- DeepSeek V4 Pro 0423 takes 1.05M tokens, gpt-oss-120b 131K.
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 V4 Pro 0423 | Bars | gpt-oss-120b | Ties |
|---|---|---|---|---|
| Coding | 7 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 2 | 0 | ||
| Reasoning | 6 | 0 | ||
| Math | 3 | 0 | ||
| Knowledge | 9 | 0 | ||
| Web design | 6 | 0 | ||
| General | 20 | 0 | ||
| All shared boards | 55 | 0 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index70
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 41.8% | Only gpt-oss-120b measured |
| SWE-bench Verified | Not measured | 26.0% | Only gpt-oss-120b measured |
| LMArena WebDev | 1447 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| AA Coding Index | 59.4 | 30.4 | DeepSeek V4 Pro 0423 |
| LiveCodeBench (Vals AI) | 87.5% | 83.2% | DeepSeek V4 Pro 0423 |
| SWE-bench Verified (Vals AI) | 77.4% | 33.6% | DeepSeek V4 Pro 0423 |
| Terminal-Bench 4 (Vals AI) | 11.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Vibe Code Bench (Vals AI) | 49.9% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Code migration (Vals AI) | 26.2% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| ProgramBench (Vals AI) | 0.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Terminal-Bench 2.1 (Vals AI) | 50.2% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| SkillsBench (Vals AI) | 51.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LiveCodeBench (Artificial Analysis run) | Not measured | 87.8% | Only gpt-oss-120b measured |
| SciCode | Not measured | 34.0% | Only gpt-oss-120b measured |
| Terminal-Bench Hard (Artificial Analysis run) | Not measured | 23.5% | Only gpt-oss-120b measured |
| Terminal-Bench | Not measured | 18.7% | Only gpt-oss-120b measured |
| FrontierCode | 17.6% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| WeirdML | 48.9% | 48.2% | DeepSeek V4 Pro 0423 |
| ALE-Bench | 1006.1 | 575.6 | DeepSeek V4 Pro 0423 |
| LMArena Text: coding | 1470 | 1379 | DeepSeek V4 Pro 0423 |
| LMArena Text: software and IT | 1469 | 1387 | DeepSeek V4 Pro 0423 |
| LMArena WebDev: React | 1441 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena WebDev: HTML | 1435 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 15 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 11.6 | DeepSeek V4 Pro 0423 |
| Epoch Capabilities Index | 149.1 | 139.8 | DeepSeek V4 Pro 0423 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| AA Agentic Index | 26.3 | 3.7 | DeepSeek V4 Pro 0423 |
| Design Arena: agents building web apps | 1000 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building full-stack apps | 948 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: agents building Godot games | 1059 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Finance Agent (Vals AI) | 44.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Tax Agent Bench (Vals AI) | 25.2% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Legal Research Bench (Vals AI) | 23.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Excel Modeling Benchmark (Vals AI) | 51.6% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| τ²-Bench Telecom | Not measured | 65.8% | Only gpt-oss-120b measured |
| APEX-Agents | Not measured | 4.4% | Only gpt-oss-120b measured |
| METR time horizons | Not measured | 56.6% | Only gpt-oss-120b measured |
| Vending-Bench 2 | $3,285 | -$22 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| GPQA Diamond | 89.4% | 78.5% | DeepSeek V4 Pro 0423 |
| GPQA Diamond (Artificial Analysis run) | Not measured | 78.2% | Only gpt-oss-120b measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 19.6% | Only gpt-oss-120b measured |
| GPQA Diamond (Epoch AI run) | 90.9% | 75.8% | DeepSeek V4 Pro 0423 |
| SimpleBench | Not measured | 22.1% | Only gpt-oss-120b measured |
| Chess puzzles | 19.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Mystery game puzzles | 17.0% | 2.0% | DeepSeek V4 Pro 0423 |
| LMArena Text: hard prompts | 1461 | 1364 | DeepSeek V4 Pro 0423 |
| LMArena Text: expert | 1463 | 1356 | DeepSeek V4 Pro 0423 |
| LMArena Text: hard prompts in English | 1466 | 1368 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 11 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| AIME | Not measured | 92.6% | Only gpt-oss-120b measured |
| MATH 500 (Vals AI) | Not measured | 94.8% | Only gpt-oss-120b measured |
| MGSM (Vals AI) | Not measured | 92.0% | Only gpt-oss-120b measured |
| AA Math Index | Not measured | 93.4 | Only gpt-oss-120b measured |
| AIME 2025 (Artificial Analysis run) | Not measured | 93.4% | Only gpt-oss-120b measured |
| OTIS Mock AIME | 96.7% | 88.9% | DeepSeek V4 Pro 0423 |
| FrontierMath, tiers 1 to 3 | 45.3% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| FrontierMath, tier 4 | 2.4% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| ProofBench | 16.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: math | 1443 | 1389 | DeepSeek V4 Pro 0423 |
| LMArena Text: mathematical work | 1455 | 1386 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 2 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index90
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| MMLU Pro | 87.3% | 79.2% | DeepSeek V4 Pro 0423 |
| LegalBench (Vals AI) | 80.3% | 75.9% | DeepSeek V4 Pro 0423 |
| CaseLaw (Vals AI) | 59.4% | 48.8% | DeepSeek V4 Pro 0423 |
| CorpFin (Vals AI) | 61.4% | 58.2% | DeepSeek V4 Pro 0423 |
| TaxEval (Vals AI) | 72.1% | 71.6% | DeepSeek V4 Pro 0423 |
| MedQA (Vals AI) | Not measured | 91.4% | Only gpt-oss-120b measured |
| MedCode (Vals AI) | 40.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| MMLU Pro (Artificial Analysis run) | Not measured | 80.8% | Only gpt-oss-120b measured |
| SimpleQA Verified | 47.0% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text: science | 1470 | 1382 | DeepSeek V4 Pro 0423 |
| LMArena Text: medicine and healthcare | 1466 | 1388 | DeepSeek V4 Pro 0423 |
| LMArena Text: legal and government | 1464 | 1355 | DeepSeek V4 Pro 0423 |
| LMArena Text: business and finance | 1443 | 1359 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 4 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| Design Arena: websites | 1235 | 968 | DeepSeek V4 Pro 0423 |
| Design Arena: UI components | 1221 | 927 | DeepSeek V4 Pro 0423 |
| Design Arena: game dev | 1229 | 992 | DeepSeek V4 Pro 0423 |
| Design Arena: data viz | 1203 | 987 | DeepSeek V4 Pro 0423 |
| Design Arena: 3D | 1254 | 901 | DeepSeek V4 Pro 0423 |
| Design Arena: all code categories | 1240 | 963 | DeepSeek V4 Pro 0423 |
| Design Arena: SVG | 1139 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Design Arena: ASCII art | 1148 | Not measured | Only DeepSeek V4 Pro 0423 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index200
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | Leader |
|---|---|---|---|
| Public Benefits Bench (Vals AI) | 62.9% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| MedScribe (Vals AI) | 75.1% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| IFBench | Not measured | 69.0% | Only gpt-oss-120b measured |
| Long context reasoning (AA-LCR) | Not measured | 52.0% | Only gpt-oss-120b measured |
| DTBench | 90.7% | 76.3% | DeepSeek V4 Pro 0423 |
| LMCA | 41.2% | 22.1% | DeepSeek V4 Pro 0423 |
| Fiction.LiveBench | Not measured | 44.4% | Only gpt-oss-120b measured |
| Creative writing (Lech Mazur) | Not measured | 77.1% | Only gpt-oss-120b measured |
| Surface Evolver Bench | 40.0% | 25.0% | DeepSeek V4 Pro 0423 |
| CL-bench Life | 13.5% | Not measured | Only DeepSeek V4 Pro 0423 measured |
| LMArena Text | 1451 | 1366 | DeepSeek V4 Pro 0423 |
| LMArena Text: instruction following | 1445 | 1318 | DeepSeek V4 Pro 0423 |
| LMArena Text: creative writing | 1446 | 1274 | DeepSeek V4 Pro 0423 |
| LMArena Text: English | 1459 | 1375 | DeepSeek V4 Pro 0423 |
| LMArena Text: non-English | 1439 | 1351 | DeepSeek V4 Pro 0423 |
| LMArena Text: Chinese | 1481 | 1385 | DeepSeek V4 Pro 0423 |
| LMArena Text: German | 1458 | 1353 | DeepSeek V4 Pro 0423 |
| LMArena Text: Japanese | 1446 | 1331 | DeepSeek V4 Pro 0423 |
| LMArena Text: Korean | 1409 | 1282 | DeepSeek V4 Pro 0423 |
| LMArena Text: French | 1473 | 1370 | DeepSeek V4 Pro 0423 |
| LMArena Text: Spanish | 1457 | 1389 | DeepSeek V4 Pro 0423 |
| LMArena Text: Russian | 1453 | 1343 | DeepSeek V4 Pro 0423 |
| LMArena Text: Polish | 1461 | 1353 | DeepSeek V4 Pro 0423 |
| LMArena Text: multi-turn | 1468 | 1340 | DeepSeek V4 Pro 0423 |
| LMArena Text: longer queries | 1458 | 1319 | DeepSeek V4 Pro 0423 |
| LMArena Text: writing and language | 1447 | 1306 | DeepSeek V4 Pro 0423 |
| LMArena Text: entertainment and media | 1431 | 1292 | DeepSeek V4 Pro 0423 |
| Neither model has a result on the other 19 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | DeepSeek V4 Pro 0423 | gpt-oss-120b | 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-oss-120b costs less
Repository review
50K input + 3K output tokens
gpt-oss-120b costs less
Agent loop
200K input + 10K output tokens
gpt-oss-120b 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 V4 Pro 0423 | OpenAI gpt-oss-120b |
|---|---|---|
| Maker | DeepSeek | OpenAI |
| Weights | Open weights | Open weights |
| Released | April 24, 2026 | August 5, 2025 |
| Context window | 1.05M | 131K |
| Input $/Mtok | $0.9553 | $0.15 |
| Output $/Mtok | $1.9105 | $0.60 |
| Coding index | 35.3 | 34.5 |
| Independent measurements behind it | 1 | 2 |
| Cost per solved task | $0.57 | $0.51 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V4 Pro 0423 better than gpt-oss-120b for coding?
DeepSeek V4 Pro 0423 and gpt-oss-120b are level for coding, 35.3 against 34.5 on the Vibe Coding Index, so price and context decide. Pick gpt-oss-120b when output tokens cost 3.2x less, or when a solved task costs $0.51 against $0.57. Pick DeepSeek V4 Pro 0423 when the context window is larger, 1.05M against 131K tokens, or when what matters is AA Coding Index or LiveCodeBench (Vals AI), where it leads.
Which is cheaper, DeepSeek V4 Pro 0423 or gpt-oss-120b?
DeepSeek V4 Pro 0423 costs $0.9553 per million input tokens and $1.9105 per million output tokens; gpt-oss-120b costs $0.15 and $0.60. Per solved SWE-bench Verified task, DeepSeek V4 Pro 0423 costs $0.57 and gpt-oss-120b $0.51.
Which has the bigger context window, DeepSeek V4 Pro 0423 or gpt-oss-120b?
DeepSeek V4 Pro 0423, with 1.05M tokens against 131K.
How many benchmarks is the DeepSeek V4 Pro 0423 vs gpt-oss-120b comparison based on?
55 benchmarks have a result for both models as of October 21, 2026. DeepSeek V4 Pro 0423's coding index rests on 1 independent benchmark and gpt-oss-120b's on 2. Where only one model has a result, the page says so instead of naming a winner.