DeepSeek V3.2 vs Kimi K2.7 Code
Kimi K2.7 Code leads DeepSeek V3.2 on coding, 53.4 to 30.0 on the Vibe Coding Index, and DeepSeek V3.2 costs 4.5x less per output token ($0.88 vs $4.00 per million). Updated October 22, 2026; 25 benchmarks have a result for both.
DeepSeek V3.2 or Kimi K2.7 Code: which to pick
Pick Kimi K2.7 Code for coding: it scores 53.4 on the Vibe Coding Index against 30.0 for DeepSeek V3.2.
Pick DeepSeek V3.2 for output tokens 4.5x cheaper ($0.88 vs $4.00 per million) and work measured by τ²-Bench Telecom, where it leads.
- Shared results
- 25
- DeepSeek V3.2 only
- 50
- Kimi K2.7 Code only
- 26
- 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
Kimi K2.7 Code
Kimi K2.7 Code scores 53.4 against 30.0.
3 and 4 benchmarks behind the scores
Intelligence
AA Intelligence Index
Kimi K2.7 Code
Kimi K2.7 Code leads, 25.8 against 21.5.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only Kimi K2.7 Code has a AA Agentic Index result, 21.0.
Directional only
Reasoning
GPQA Diamond
No call
Only DeepSeek V3.2 has a GPQA Diamond result, 80.3%.
Directional only
Math
AIME
No call
Only DeepSeek V3.2 has a AIME result, 84.6%.
Directional only
Knowledge
MMLU Pro
No call
Only DeepSeek V3.2 has a MMLU Pro result, 84.9%.
Directional only
Web design
Design Arena: websites
Kimi K2.7 Code
Kimi K2.7 Code leads, 1263 against 1175.
Same board, both measured
Long documents
Listed context window
Kimi K2.7 Code
Kimi K2.7 Code takes 262K tokens against 164K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
DeepSeek V3.2
DeepSeek V3.2 $0.0007, Kimi K2.7 Code $0.0029 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
Kimi K2.7 Code
DeepSeek V3.2 $0.99, Kimi K2.7 Code $0.35.
Published attempt cost and pass rate
At a glance
- Coding index
- DeepSeek V3.2 30.0 from 3 benchmarks, Kimi K2.7 Code 53.4 from 4 benchmarks.
- Widest category
- Coding, 16.2 points apart on average over 8 shared benchmarks.
- Widest single gap
- Vibe Code Bench (Vals AI): DeepSeek V3.2 5.1% against Kimi K2.7 Code 47.2%, 42.1 points apart on a 0 to 100 scale.
- Output price
- DeepSeek V3.2 $0.88 against Kimi K2.7 Code $4.00 per million output tokens, 4.5x apart.
- Context window
- DeepSeek V3.2 takes 164K tokens, Kimi K2.7 Code 262K.
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 V3.2 | Bars | Kimi K2.7 Code | Ties |
|---|---|---|---|---|
| Coding | 0 | 8 | ||
| Intelligence | 0 | 2 | ||
| Agentic work | 1 | 2 | ||
| Reasoning | 0 | 2 | ||
| Web design | 0 | 8 | ||
| General | 0 | 2 | ||
| All shared boards | 1 | 24 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index08
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| SWE-bench Verified | 70.0% | Not measured | Only DeepSeek V3.2 measured |
| SWE-bench Multilingual | 59.0% | Not measured | Only DeepSeek V3.2 measured |
| LMArena WebDev | 1362 | 1474 | Kimi K2.7 Code |
| AA Coding Index | 44.2 | 60.8 | Kimi K2.7 Code |
| LiveCodeBench (Vals AI) | 80.7% | 82.0% | Kimi K2.7 Code |
| SWE-bench Verified (Vals AI) | 67.6% | 78.2% | Kimi K2.7 Code |
| Vibe Code Bench (Vals AI) | 5.1% | 47.2% | Kimi K2.7 Code |
| Code migration (Vals AI) | Not measured | 25.4% | Only Kimi K2.7 Code measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only Kimi K2.7 Code measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 67.0% | Only Kimi K2.7 Code measured |
| SkillsBench (Vals AI) | Not measured | 50.0% | Only Kimi K2.7 Code measured |
| LiveCodeBench (Artificial Analysis run) | 86.2% | Not measured | Only DeepSeek V3.2 measured |
| SciCode | Not measured | 47.8% | Only Kimi K2.7 Code measured |
| Terminal-Bench Hard (Artificial Analysis run) | 35.6% | 44.7% | Kimi K2.7 Code |
| Terminal-Bench | 39.6% | Not measured | Only DeepSeek V3.2 measured |
| FrontierCode | Not measured | 30.1% | Only Kimi K2.7 Code measured |
| WeirdML | Not measured | 54.1% | Only Kimi K2.7 Code measured |
| ALE-Bench | Not measured | 886.2 | Only Kimi K2.7 Code measured |
| LMArena Text: coding | 1453 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: software and IT | 1449 | Not measured | Only DeepSeek V3.2 measured |
| LMArena WebDev: React | 1350 | 1475 | Kimi K2.7 Code |
| LMArena WebDev: HTML | 1371 | 1451 | Kimi K2.7 Code |
| LMArena WebDev: from an image | Not measured | 1513 | Only Kimi K2.7 Code measured |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| AA Intelligence Index | 21.5 | 25.8 | Kimi K2.7 Code |
| Epoch Capabilities Index | 146.1 | 150.1 | Kimi K2.7 Code |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index12
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 21.0 | Only Kimi K2.7 Code measured |
| Design Arena: agents building web apps | Not measured | 1176 | Only Kimi K2.7 Code measured |
| Design Arena: agents building full-stack apps | Not measured | 1175 | Only Kimi K2.7 Code measured |
| Design Arena: agents building mobile apps | Not measured | 1143 | Only Kimi K2.7 Code measured |
| Design Arena: agents building Android apps | Not measured | 1138 | Only Kimi K2.7 Code measured |
| Design Arena: agents building Godot games | Not measured | 1147 | Only Kimi K2.7 Code measured |
| Design Arena: agents building games | Not measured | 1130 | Only Kimi K2.7 Code measured |
| Design Arena: HTML slides | Not measured | 1151 | Only Kimi K2.7 Code measured |
| Design Arena: PowerPoint slides | Not measured | 1129 | Only Kimi K2.7 Code measured |
| τ²-Bench Telecom | 90.6% | 90.1% | DeepSeek V3.2 |
| APEX-Agents | 21.3% | 37.6% | Kimi K2.7 Code |
| Vending-Bench 2 | $1,034 | $5,083 | Kimi K2.7 Code |
| Neither model has a result on the other 29 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| GPQA Diamond | 80.3% | Not measured | Only DeepSeek V3.2 measured |
| GPQA Diamond (Artificial Analysis run) | 84.0% | 89.6% | Kimi K2.7 Code |
| Humanity's Last Exam (Artificial Analysis run) | 24.6% | 35.0% | Kimi K2.7 Code |
| GPQA Diamond (Epoch AI run) | Not measured | 87.9% | Only Kimi K2.7 Code measured |
| ARC-AGI | 57.0% | Not measured | Only DeepSeek V3.2 measured |
| ARC-AGI-2 | 4.0% | Not measured | Only DeepSeek V3.2 measured |
| SimpleBench | Not measured | 57.9% | Only Kimi K2.7 Code measured |
| Chess puzzles | Not measured | 14.0% | Only Kimi K2.7 Code measured |
| LMArena Text: hard prompts | 1434 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: expert | 1436 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: hard prompts in English | 1446 | Not measured | Only DeepSeek V3.2 measured |
| Neither model has a result on the other 11 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index9 boards, none shared
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| AIME | 84.6% | Not measured | Only DeepSeek V3.2 measured |
| MGSM (Vals AI) | 86.0% | Not measured | Only DeepSeek V3.2 measured |
| AA Math Index | 92.0 | Not measured | Only DeepSeek V3.2 measured |
| AIME 2025 (Artificial Analysis run) | 92.0% | Not measured | Only DeepSeek V3.2 measured |
| OTIS Mock AIME | Not measured | 95.6% | Only Kimi K2.7 Code measured |
| FrontierMath, tiers 1 to 3 | Not measured | 54.0% | Only Kimi K2.7 Code measured |
| FrontierMath, tier 4 | Not measured | 12.2% | Only Kimi K2.7 Code measured |
| LMArena Text: math | 1435 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: mathematical work | 1440 | Not measured | Only DeepSeek V3.2 measured |
| Neither model has a result on the other 4 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index12 boards, none shared
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| MMLU Pro | 84.9% | Not measured | Only DeepSeek V3.2 measured |
| LegalBench (Vals AI) | 76.1% | Not measured | Only DeepSeek V3.2 measured |
| CaseLaw (Vals AI) | 55.4% | Not measured | Only DeepSeek V3.2 measured |
| CorpFin (Vals AI) | 51.0% | Not measured | Only DeepSeek V3.2 measured |
| TaxEval (Vals AI) | 68.2% | Not measured | Only DeepSeek V3.2 measured |
| MedQA (Vals AI) | 93.9% | Not measured | Only DeepSeek V3.2 measured |
| MMLU Pro (Artificial Analysis run) | 86.2% | Not measured | Only DeepSeek V3.2 measured |
| SimpleQA Verified | Not measured | 36.5% | Only Kimi K2.7 Code measured |
| LMArena Text: science | 1446 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: medicine and healthcare | 1440 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: legal and government | 1425 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: business and finance | 1410 | Not measured | Only DeepSeek V3.2 measured |
| Neither model has a result on the other 6 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index08
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| Design Arena: websites | 1175 | 1263 | Kimi K2.7 Code |
| Design Arena: UI components | 1146 | 1255 | Kimi K2.7 Code |
| Design Arena: game dev | 1131 | 1220 | Kimi K2.7 Code |
| Design Arena: data viz | 1158 | 1227 | Kimi K2.7 Code |
| Design Arena: 3D | 1132 | 1240 | Kimi K2.7 Code |
| Design Arena: all code categories | 1163 | 1252 | Kimi K2.7 Code |
| Design Arena: SVG | 1032 | 1176 | Kimi K2.7 Code |
| Design Arena: ASCII art | 1083 | 1196 | Kimi K2.7 Code |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index02
| Benchmark | DeepSeek V3.2 | Kimi K2.7 Code | Leader |
|---|---|---|---|
| IFBench | 60.7% | 63.1% | Kimi K2.7 Code |
| Long context reasoning (AA-LCR) | 73.3% | 79.3% | Kimi K2.7 Code |
| DTBench | 85.6% | Not measured | Only DeepSeek V3.2 measured |
| LMCA | 28.8% | Not measured | Only DeepSeek V3.2 measured |
| CL-bench | 12.4% | Not measured | Only DeepSeek V3.2 measured |
| Surface Evolver Bench | Not measured | 48.8% | Only Kimi K2.7 Code measured |
| CL-bench Life | 7.4% | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text | 1425 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: instruction following | 1413 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: creative writing | 1400 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: English | 1439 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: non-English | 1409 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Chinese | 1455 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: German | 1413 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Japanese | 1374 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Korean | 1370 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: French | 1429 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Spanish | 1424 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Russian | 1424 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: Polish | 1419 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: multi-turn | 1427 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: longer queries | 1428 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: writing and language | 1407 | Not measured | Only DeepSeek V3.2 measured |
| LMArena Text: entertainment and media | 1395 | Not measured | Only DeepSeek V3.2 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 | DeepSeek V3.2 | Kimi K2.7 Code | 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 V3.2 costs less
Repository review
50K input + 3K output tokens
DeepSeek V3.2 costs less
Agent loop
200K input + 10K output tokens
DeepSeek V3.2 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 V3.2 | MoonshotAI Kimi K2.7 Code |
|---|---|---|
| Maker | DeepSeek | Moonshot AI |
| Weights | Open weights | Open weights |
| Released | December 1, 2025 | June 12, 2026 |
| Context window | 164K | 262K |
| Input $/Mtok | $0.29 | $0.95 |
| Output $/Mtok | $0.88 | $4.00 |
| Coding index | 30.0 | 53.4 |
| Independent measurements behind it | 3 | 4 |
| Cost per solved task | $0.99 | $0.35 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is DeepSeek V3.2 better than Kimi K2.7 Code for coding?
Pick Kimi K2.7 Code for coding: it scores 53.4 on the Vibe Coding Index against 30.0 for DeepSeek V3.2. Pick DeepSeek V3.2 for output tokens 4.5x cheaper ($0.88 vs $4.00 per million) and work measured by τ²-Bench Telecom, where it leads.
Which is cheaper, DeepSeek V3.2 or Kimi K2.7 Code?
DeepSeek V3.2 costs $0.29 per million input tokens and $0.88 per million output tokens; Kimi K2.7 Code costs $0.95 and $4.00. Per solved SWE-bench Verified task, DeepSeek V3.2 costs $0.99 and Kimi K2.7 Code $0.35.
Which has the bigger context window, DeepSeek V3.2 or Kimi K2.7 Code?
Kimi K2.7 Code, with 262K tokens against 164K.
How many benchmarks is the DeepSeek V3.2 vs Kimi K2.7 Code comparison based on?
25 benchmarks have a result for both models as of October 22, 2026. DeepSeek V3.2's coding index rests on 3 independent benchmarks and Kimi K2.7 Code's on 4. Where only one model has a result, the page says so instead of naming a winner.