ResearchLLM Verifications Leaderboard

Data as of · 78 models, 80 questions · reviewed by Suhas Bhairav · methodology

Head to head · tested 27 Oct 2026

Kimi K2.7 Code vs Kimi K3.1 Pro: accuracy, speed and price

Similar accuracy; Kimi K2.7 Code costs less per question.

MoonshotAIKimi K2.7 Code65.0% score · #33$0.006 per question136 tok/sMoonshotAIKimi K3.1 Pro77.5% score · #9$0.01 per question38 tok/s

Verdict

Built from the measurements below

  • Accuracy is too close to call: Kimi K2.7 Code scored 65.0% and Kimi K3.1 Pro 77.5% on the same 80 questions, a gap within what chance alone could produce.
  • Kimi K2.7 Code is 1.7× cheaper per question as billed ($0.006 vs $0.01), reasoning tokens included.
  • Kimi K2.7 Code generates text 3.6× faster: 136 tokens per second against 38.
  • Kimi K2.7 Code starts answering sooner: 0.4 s to the first token, against 1.3 s for Kimi K3.1 Pro.
  • Kimi K3.1 Pro accepts more context: 1M tokens against 262k.

Which should you choose?

Choose Kimi K2.7 Code for the lower cost per correct answer ($0.0092 against $0.014), faster streaming, a quicker first token, open weights you can run yourself and more hosts to choose from on OpenRouter (11 against 1). Choose Kimi K3.1 Pro for shorter outputs (about 973 tokens per question against 1,781) and a 1M-token context window. The widest subject gap is in engineering: Kimi K3.1 Pro answered 11 of 15 correctly, Kimi K2.7 Code 7 of 15. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap -12.5 points (z = -1.75). A winner is called only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.

Head to head

Same questions, same prompt, same API

MeasureKimi K2.7 CodeKimi K3.1 ProEdge
Verification scoreHigher is better65.0%77.5%Too close to call
Cost per questionLower is better$0.006$0.01Kimi K2.7 Code
Output speedHigher is better13638Kimi K2.7 Code
Time to first tokenLower is better0.4 s1.3 sKimi K2.7 Code
End-to-end response timeLower is better5.3 s19.6 sKimi K2.7 Code
Tokens per questionLower is better1,781973Kimi K3.1 Pro
Context windowHigher is better262k1MKimi K3.1 Pro

Side by side

Bars are scaled to the larger of the two values. The winner on each measure is marked in bold.

Every measure

Kimi K2.7 Code first, Kimi K3.1 Pro second

Verification scorehigher is better
Kimi K2.7 Code65.0%
Kimi K3.1 Pro77.5%
Cost per questionlower is better
Kimi K2.7 Code$0.006
Kimi K3.1 Pro$0.01
Output speedhigher is better
Kimi K2.7 Code136
Kimi K3.1 Pro38
Time to first tokenlower is better
Kimi K2.7 Code0.4 s
Kimi K3.1 Pro1.3 s
End-to-end response timelower is better
Kimi K2.7 Code5.3 s
Kimi K3.1 Pro19.6 s
Tokens per questionlower is better
Kimi K2.7 Code1,781
Kimi K3.1 Pro973
Context windowhigher is better
Kimi K2.7 Code262k
Kimi K3.1 Pro1M

Score vs. cost per question

Both models in color, every other model in grey

Most attractive quadrant0255075100$0.00001$0.0001$0.001$0.01$0.10Cost per question (USD, log) →Verification score (%) →Qwen3.8 2.4T A95BCommand AReka Flash 3Gemma 4 26B A4BGemma 4 31BQwen3.5 397B A17BHermes 4 405BMuse WonderGrok 4.7o3gpt-oss-20bMinistral 3 3B 2512Nova Micro 1.0Phi 4Granite 4.2 8BHy3GLM 5.3 FlashSolar Pro 4Trinity Large ThinkingMiniMax M3Mistral Large 4Claude Opus 5.5GPT-6.1 SolGemini 4 ArgonGPT-5.4GPT-6 Lunagpt-oss-120bLlama 4 Maverick

Specifications and pricing

From OpenRouter's catalogue, plus the leaderboard's test dates

SpecificationKimi K2.7 CodeKimi K3.1 Pro
CreatorMoonshotAIMoonshotAI
Released12 Jun 202623 Oct 2026
WeightsOpen weights—
ReasoningYesYes
Inputtext, imagetext, image, video
Outputtexttext
Context window262k tokens1M tokens
Knowledge cutoff——
List price in / out per 1M$0.56 / $3.1$2 / $10
Blended price per 1M (3:1)$1.20$4.00
Cheapest hostCoreWeave · $0.6 / $3Moonshot AI · $2 / $10
Hosts on OpenRouter111
Last tested27 Oct 202625 Oct 2026

Compare any two models →

Kimi K2.7 Code vs Kimi K3.1 Pro: frequently asked questions

Is Kimi K2.7 Code better than Kimi K3.1 Pro?
On accuracy they are statistically tied: Kimi K2.7 Code scored 65.0% and Kimi K3.1 Pro 77.5% on the same 80 questions from open datasets. Similar accuracy; Kimi K2.7 Code costs less per question.
Which is cheaper, Kimi K2.7 Code or Kimi K3.1 Pro?
Kimi K2.7 Code is cheaper: $0.006 per question as billed, against $0.01. List prices: Kimi K2.7 Code $0.56 in / $3.1 out, Kimi K3.1 Pro $2 in / $10 out per million tokens.
Which is faster, Kimi K2.7 Code or Kimi K3.1 Pro?
Kimi K2.7 Code is faster at 136 output tokens per second, against 38. Time to first token: 0.4 s for Kimi K2.7 Code, 1.3 s for Kimi K3.1 Pro.
Which has the larger context window, Kimi K2.7 Code or Kimi K3.1 Pro?
Kimi K3.1 Pro, with 1M tokens against 262k.