Model comparison
DeepSeek-V3.2-Exp vs Kimi K2.7 Code
DeepSeek-V3.2-Exp and Kimi K2.7 Code score almost the same on the Noometry Index (44.3 vs 43.3), so choose on price, context window or the category you care about most.
Last verified . 10 shared benchmarks.
DeepSeekSummary
- They share 10 benchmarks with published results for both. DeepSeek-V3.2-Exp scores higher in 2 categories and Kimi K2.7 Code in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Kimi K2.7 Code leads 39.0 to 22.1.
- The biggest single-benchmark swing is APEX-Agents: 21.3% for DeepSeek-V3.2-Exp and 37.6% for Kimi K2.7 Code.
- DeepSeek-V3.2-Exp is cheaper at $0.26 / $0.38 per million input/output tokens, against $0.95 / $4 for Kimi K2.7 Code.
- Kimi K2.7 Code accepts more context: 262K tokens versus 164K.
Side by side
| DeepSeek-V3.2-Exp | Kimi K2.7 Code | |
|---|---|---|
| Provider | DeepSeek | Moonshot AI |
| Noometry Index | 44.3 | 43.3 |
| Released | 2025-09-29 | 2026-06-12 |
| Weights | Open | Open |
| Context window | 164K | 262K |
| Max output | 66K | 262K |
| Input $ / M tokens | $0.26 | $0.95 |
| Output $ / M tokens | $0.38 | $4 |
| Results tracked | 49 | 19 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 46.5 (#65), Kimi K2.7 Code: 42.9 (#95)
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| LMArena WebDev | 1362 | 1473 |
| SciCode | 38.9% | 47.5% |
| WeirdML | 39.5% | 54.1% |
| DeepSWE | — | 30.5% |
| FrontierCode | — | 30.1% |
| SWE-bench Verified (bash only) | 70% | — |
| Aider Polyglot | 74.2% | — |
| SWE-bench Multilingual | 59% | — |
| LMArena Coding | 1454 | — |
| ALE-Bench | — | 886.23 |
Agentic & Tool Use DeepSeek-V3.2-Exp leads
DeepSeek-V3.2-Exp: 32.7 (#59), Kimi K2.7 Code: 24.0 (#122)
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| APEX-Agents | 21.3% | 37.6% |
| Vending-Bench 2 | 1,034 | 5,083 |
| Terminal-Bench | 39.6% | — |
| Berkeley Function Calling Leaderboard | 56.7% | — |
| TheAgentCompany | 42.9% | — |
| GBAEval | — | 0.9% |
Reasoning Kimi K2.7 Code leads
DeepSeek-V3.2-Exp: 22.1 (#208), Kimi K2.7 Code: 39.0 (#61)
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| CritPt | 2.9% | 10% |
| Chess Puzzles | 14% | 21% |
| Epoch Capabilities Index | 146.27 | 149.97 |
| ARC-AGI-2 | 4% | — |
| SimpleBench | — | 57.9% |
| Kagi LLM Benchmark | 52.2% | — |
| NYT Connections (extended) | 36.7% | — |
| ARC-AGI-1 | 57% | — |
| Thematic Generalization | 65% | — |
| LMArena Hard Prompts | 1434 | — |
| DTBench | 87.7% | — |
| LMCA | 29.1% | — |
| Surface Evolver Bench | — | 48.8% |
Math Kimi K2.7 Code leads
DeepSeek-V3.2-Exp: 41.7 (#87), Kimi K2.7 Code: 52.9 (#48)
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 87.8% | 95.6% |
| FrontierMath (Tiers 1-3) | — | 54% |
| FrontierMath Tier 4 | — | 12.2% |
| MathArena Final-Answer Competitions | 57.7% | — |
| ProofBench | 8% | — |
| LMArena Math | 1435 | — |
| FrontierMath (Feb 2025 set) | 22.1% | — |
| FrontierMath Tier 4 (v1) | 2.1% | — |
Knowledge Kimi K2.7 Code leads
DeepSeek-V3.2-Exp: 51.7 (#66), Kimi K2.7 Code: 53.5 (#57)
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| GPQA Diamond | 83.4% | 87.9% |
| SimpleQA Verified | — | 36.5% |
| Vectara Hallucination Rate | 5.3% | — |
| LMArena Expert | 1436 | — |
Multilingual Not comparable
DeepSeek-V3.2-Exp: 52.2 (#90), Kimi K2.7 Code: —
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| LMArena Non-English | 1409 | — |
| LMArena Chinese | 1461 | — |
| LMArena French | 1433 | — |
| LMArena German | 1440 | — |
| LMArena Japanese | 1374 | — |
| LMArena Korean | 1371 | — |
| LMArena Russian | 1424 | — |
| LMArena Spanish | 1440 | — |
Instruction Following Not comparable
DeepSeek-V3.2-Exp: 74.5 (#93), Kimi K2.7 Code: —
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| LMArena Instruction Following | 1413 | — |
Long Context Not comparable
DeepSeek-V3.2-Exp: 47.6 (#16), Kimi K2.7 Code: —
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| Fiction.LiveBench | 83.3% | — |
| CL-bench | 13.2% | — |
| CL-bench Life | 9.5% | — |
| LMArena Longer Query | 1428 | — |
Writing & Preference Not comparable
DeepSeek-V3.2-Exp: 62.4 (#77), Kimi K2.7 Code: —
| Benchmark | DeepSeek-V3.2-Exp | Kimi K2.7 Code |
|---|---|---|
| LMArena Text | 1425 | — |
| LMArena Creative Writing | 1403 | — |
| EQ-Bench Creative Writing | 1515 | — |
| LMArena Multi-Turn | 1427 | — |
Frequently asked questions
Is DeepSeek-V3.2-Exp better than Kimi K2.7 Code?
DeepSeek-V3.2-Exp and Kimi K2.7 Code score almost the same on the Noometry Index (44.3 vs 43.3), so choose on price, context window or the category you care about most.
Which is cheaper, DeepSeek-V3.2-Exp or Kimi K2.7 Code?
DeepSeek-V3.2-Exp is cheaper. It lists at $0.26 per million input tokens and $0.38 per million output tokens; Kimi K2.7 Code lists at $0.95 and $4.
Is DeepSeek-V3.2-Exp or Kimi K2.7 Code better for coding?
DeepSeek-V3.2-Exp scores higher on coding benchmarks: 46.5 versus 42.9 in the Noometry coding category.
Which has the bigger context window?
Kimi K2.7 Code does, with 262K tokens against 164K.
How many benchmarks do DeepSeek-V3.2-Exp and Kimi K2.7 Code share?
10 benchmarks have published results for both models. DeepSeek-V3.2-Exp has 49 scored results on Noometry and Kimi K2.7 Code has 19.