Skip to content
ModelCap

Head-to-head comparison

DeepSeek V3.2 vs Kimi K2.7 Code

DeepSeek V3.2 (DeepSeek) and Kimi K2.7 Code (Moonshot AI) compared on the ModelCap Index, API price, context window, provider availability, weight access and every public benchmark board they share. Figures are the same ones shown on the live rankings; nothing here is a hidden score.

Snapshot as of 9 September 2026

As of 9 September 2026, Kimi K2.7 Code holds the stronger ModelCap Index position (#47 vs #51); DeepSeek V3.2 is 8.8× cheaper per output token ($0.40/1M vs $3.50/1M); Kimi K2.7 Code offers the longer context window (262K vs 164K tokens); and DeepSeek V3.2 ships open weights.

Which should you choose?

Choose DeepSeek V3.2 if…

  • API cost matters — $0.40/1M output tokens against $3.50/1M, about 8.8× cheaper.
  • your workload is prompt-heavy — input tokens cost $0.269/1M against $0.71/1M.
  • you want to self-host: DeepSeek V3.2 ships open weights while Kimi K2.7 Code is restricted license.
  • you want provider choice — 15 listed API providers against 14.

Choose Kimi K2.7 Code if…

  • you want the stronger overall ModelCap Index position — #47 against #51 (59.5 vs 58.2 points).
  • you need the longer context window — 262K tokens against 164K.
  • AA Terminal-Bench 2.1 is your yardstick — 67.4% against 46.8%.
  • AA τ²-Bench Telecom is your yardstick — 90.1% against 78.9%.
  • AA Humanity's Last Exam is your yardstick — 35.0% against 11.2%.
  • WildClawBench OpenClaw is your yardstick — 46.9 against 34.0.
  • you want the more recently listed model — Kimi K2.7 Code was listed 12 June 2026, DeepSeek V3.2 1 December 2025.

DeepSeek V3.2 vs Kimi K2.7 Code: specs, pricing and context

Specification comparison of DeepSeek V3.2 and Kimi K2.7 Code
FieldDeepSeek V3.2DeepSeekKimi K2.7 CodeMoonshot AI
ModelCap Index position#51#47
Index score58.259.5
EvidenceMeasuredMeasured
Input price / 1M tokens$0.269$0.71
Output price / 1M tokens$0.40$3.50
Context window164K tokens262K tokens
Max output tokens66K236K
API providers listed1514
Weight accessOpen weightsRestricted license
Input modalitiesTextText, Image
First listed1 December 202512 June 2026
PublisherDeepSeekMoonshot AI

Evidence: 4 public benchmark observations across 4 boards · 2 public benchmark observations across 2 boards. Prices are the lowest listed API offer per million tokens observed on the OpenRouter catalogue.

Benchmark scores: DeepSeek V3.2 vs Kimi K2.7 Code

Public benchmark boards where DeepSeek V3.2 or Kimi K2.7 Code has a published result
BoardDeepSeek V3.2Kimi K2.7 CodeLeads
Arena codingArena (LMArena)14761469–1483Thinking—Only one result
ARC-AGI-2ARC Prize Foundation4.0%—Only one result
SWE-benchSWE-bench (Princeton / SWE-agent team)70.0%High—Only one result
AA Intelligence IndexArtificial Analysis—26.3intelligence-index:v4.3:reasoning-unspecifiedOnly one result
AA Terminal-Bench 2.1Artificial Analysis46.8%terminal-bench:2.1:terminus-2:e2b:pass-at-1:3-repeats:source-model="DeepSeek V3.2":reasoning=true67.4%terminal-bench:2.1:terminus-2:e2b:pass-at-1:3-repeats:source-model="Kimi K2.7 Code":reasoning=trueKimi K2.7 Code
AA τ²-Bench TelecomArtificial Analysis78.9%tau2:telecom:dual-control:pass-at-1:3-repeats:source-model="DeepSeek V3.2 (Non-reasoning)":reasoning=false90.1%tau2:telecom:dual-control:pass-at-1:3-repeats:source-model="Kimi K2.7 Code":reasoning=trueKimi K2.7 Code
AA Humanity's Last ExamArtificial Analysis11.2%hle:may-2025:text-only-2158:no-tools:pass-at-1:source-model="DeepSeek V3.2 (Non-reasoning)":reasoning=false35.0%hle:may-2025:text-only-2158:no-tools:pass-at-1:source-model="Kimi K2.7 Code":reasoning=trueKimi K2.7 Code
WildClawBench OpenClawWildClawBench34.046.9Kimi K2.7 Code

Scores are the sources' own published figures for each model's best evaluated configuration; ModelCap never re-runs a benchmark.

Want a different pairing? Open the interactive comparison tool to swap either model for any current ranked language model.

DeepSeek V3.2 vs Kimi K2.7 Code: common questions

Is DeepSeek V3.2 better than Kimi K2.7 Code?

Kimi K2.7 Code ranks higher on the ModelCap Index as of 9 September 2026: #47 against #51. That is a capability ranking built from public benchmark evidence with published uncertainty; whether it is "better" for you also depends on price, context and where you can run it. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.

Is DeepSeek V3.2 cheaper than Kimi K2.7 Code?

DeepSeek V3.2 is cheaper on output tokens: $0.40/1M against $3.50/1M. Input tokens are $0.269/1M for DeepSeek V3.2 and $0.71/1M for Kimi K2.7 Code. Prices are the lowest listed API offer ModelCap observed, in USD per million tokens. For 1,000 requests with 2,000 input and 500 output tokens each (2M input + 0.5M output), the listed-rate estimate is $0.74 for DeepSeek V3.2 versus $3.17 for Kimi K2.7 Code. DeepSeek V3.2 costs 76.7% less in this scenario. This excludes caching, batch discounts, prompt-length tiers, tool charges and retries; verify the selected endpoint before budgeting.

Which has the bigger context window, DeepSeek V3.2 or Kimi K2.7 Code?

Kimi K2.7 Code has the larger context window: 262K tokens against 164K.

Which is better for coding, DeepSeek V3.2 or Kimi K2.7 Code?

AA Terminal-Bench 2.1: DeepSeek V3.2 46.8%, Kimi K2.7 Code 67.4% — Kimi K2.7 Code leads.

Are DeepSeek V3.2 and Kimi K2.7 Code open-weight models?

DeepSeek V3.2: Open weights. Kimi K2.7 Code: Restricted license. Open weights mean the checkpoint can be downloaded and self-hosted under its licence; API-only models are available solely through hosted endpoints.

Where can I run DeepSeek V3.2 and Kimi K2.7 Code?

ModelCap currently lists 15 API providers for DeepSeek V3.2 and 14 for Kimi K2.7 Code, from the OpenRouter catalogue snapshot the site serves; each model page lists the providers and their prices.

How current is this DeepSeek V3.2 vs Kimi K2.7 Code comparison?

Every figure comes from the sealed ModelCap dataset published 9 September 2026; the page re-renders within a minute of each data refresh, and the ModelCap Index positions are the same ones shown on the live rankings.

Explore further