Compare / head-to-head
DeepSeek V3.2-Expvs
Kimi K2
DeepSeek V3.2-Exp leads 5 of 5 shared benchmarks. Both offer a 128K-token context window.
Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.
Shared benchmarks
5 – 0
DeepSeek V3.2-Exp leads
Cheaper per token
—
list price, input + output
Larger context
Tie
both 128K tokens
Providers
2 providers
DeepSeek · Moonshot AI
DeepSeek V3.2-Exp
DeepSeek · released 2025-09-29
text
- Context
- 128K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 17 · 7 core
Kimi K2
Moonshot AI · released 2025-07-11
text
- Context
- 128K
- Max out
- —
- Input /1M
- —
- Output /1M
- —
- Cached /1M
- —
- Scores
- 18 · 7 core
Quality
Benchmark matrix
5 shared · 12 only DeepSeek V3.2-Exp · 13 only Kimi K2| Benchmark | DeepSeek V3.2-Exp | Kimi K2 | Δ | Edge |
|---|---|---|---|---|
| Reported by both · 5 | ||||
| MMLU-Pro | 85% ↗ | 81.1% ↗ | +3.9 pt | DeepSeek V3.2-Exp |
| AIME 2025 | 89.3% ↗ | 49.5% ↗ | +39.8 pt | DeepSeek V3.2-Exp |
| Aider-Polyglot | 74.5% ↗ | 60% ↗ | +14.5 pt | DeepSeek V3.2-Exp |
| GPQA-Diamond | 79.9% ↗ | 75.1% ↗ | +4.8 pt | DeepSeek V3.2-Exp |
| WeirdML (Håvard Tveit Ihle) | 39.5% ↗ | 39.4% ↗ | +0.1 pt | DeepSeek V3.2-Exp |
| Only DeepSeek V3.2-Exp reports · 12 | ||||
| Humanity's Last Exam (no tools) | 19.8% ↗ | not reported | — | — |
| LMArena Elo | 1424.9 ↗ | not reported | — | — |
| BrowseComp | 40.1% ↗ | not reported | — | — |
| BrowseComp-zh | 47.9% ↗ | not reported | — | — |
| HMMT 2025 | 83.6% ↗ | not reported | — | — |
| LiveCodeBench | 74.1% ↗ | not reported | — | — |
| LMArena WebDev (LMArena) | 1271.8 ↗ | not reported | — | — |
| SimpleQA (agentic tool use) | 97.1% ↗ | not reported | — | — |
| SWE Verified | 67.8% ↗ | not reported | — | — |
| SWE-bench Multilingual | 57.9% ↗ | not reported | — | — |
| Terminal-bench | 37.7% ↗ | not reported | — | — |
| Vending-Bench 2 (Andon Labs) | 1034 ↗ | not reported | — | — |
| Only Kimi K2 reports · 13 | ||||
| GSO Opt@1 (GSO) | not reported | 2% ↗ | — | — |
| Humanity's Last Exam (Text Only) | not reported | 4.7% ↗ | — | — |
| IFEval (Prompt Strict) | not reported | 89.8% ↗ | — | — |
| LiveCodeBench v6 (Aug 24 - May 25) | not reported | 53.7% ↗ | — | — |
| MMLU-Redux | not reported | 92.7% ↗ | — | — |
| SimpleBench (SimpleBench) | not reported | 26.3% ↗ | — | — |
| SimpleQA | not reported | 31% ↗ | — | — |
| SWE-bench Multilingual (Agentic Coding, single attempt) | not reported | 47.3% ↗ | — | — |
| SWE-bench Verified (Agentic Coding, single attempt) | not reported | 65.8% ↗ | — | — |
| Tau2 airline | not reported | 56.5% ↗ | — | — |
| Tau2 retail | not reported | 70.6% ↗ | — | — |
| Tau2 telecom | not reported | 65.8% ↗ | — | — |
| TerminalBench (Inhouse Framework) | not reported | 30% ↗ | — | — |
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is DeepSeek V3.2-Exp minus Kimi K2 in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing
Side by side
Official pricing pages and model cards| Spec | DeepSeek V3.2-Exp | Kimi K2 | Edge |
|---|---|---|---|
| Context window | 128K tokens | 128K tokens | Tie |
| Max output | — | — | — |
| Input price / 1M | — | — | — |
| Output price / 1M | — | — | — |
| Cached input / 1M | — | — | — |
| Input + output / 1M Lower is cheaper. List prices; batch, tool and regional fees excluded. | — | — | — |
| Modalities | text | text | Tie |
| Released | 2025-09-29 | 2025-07-11 | — |
| Cited benchmark scores | 17 | 18 | — |
Reliability
Provider status
Live from /statusMore matchups
DeepSeek V3.2-Exp vs …
Models sharing the most benchmarksMore matchups
Kimi K2 vs …
Models sharing the most benchmarksBuilt by Respan
Which one wins on your data?
Public benchmarks are a starting point. Run DeepSeek V3.2-Exp and Kimi K2 on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.