Kimi K2.5

Moonshot AIOpen weightsModified MITReleased Jan 27, 2026

Kimi K2.5 is not on the arena.ai Agents leaderboard, so it has no rank on our default leaderboard; the other data below is for reference. Kimi K2.5 ranks #81 on the LMArena Text leaderboard (style control) with an Arena score of 1449.5 (±3.3, 73,872 votes). API price: $0.60 input / $3.00 output per 1M tokens. Context: 262K. Nearest rivals: GPT-5.6 Luna (1453.2) above and Gemini 2.5 Pro (1445.5) below.

Quick answer

Agents leaderboard
No result yet
Text (LMArena)
#81 · 1449.5
API price (in / out)
$0.60 / $3.00 per 1M tokens
Data updated
Nov 10, 2026 · synced daily

Sources: arena.ai, LMArena, Artificial Analysis, OpenRouter

Key metrics

Agent Net Improvement–arena.ai Agents
Arena score1449.5±3.3LMArena #81 · 73,872 votes
Intelligence–Variant: reasoning
Coding–Artificial Analysis
Writing (creative)1421.7±6.2LMArena creative writing #67 / 419
List price in / out$0.60 / $3.00USD per 1M tokens
Lowest route$0.53 / $3.33Cheapest OpenRouter route
Context262Ktokens
Output speed–AA measured median
TTFT / first answer–AA measured median

Rank history

LMArena rank, last 26 days
#77#80#82Oct 16Nov 10

#79 on Oct 16, #81 now; best #78 in this period.

Models ranked nearby

Sorted by Arena score
#ModelArenaIntel.Output $tok/s
51Qwen3.7 Plus1455.3–$1.6058
52Hy31454.2–$0.5692
53GPT-5.6 Luna1453.223.8$1.20–
54Kimi K2.5This page1449.5–$3.00–
55Gemini 2.5 Pro1445.5–$10.00–
56Grok 4.71442.035.5$6.0058
57Qwen3.5 397B1441.5–$3.60–

More from Moonshot AI

#ModelArenaIntel.Output $tok/s
16Kimi K3.1 Pro1488.046.5$10.0028
17Kimi K31487.117.4$15.00–
45Kimi K2.61460.8–$4.00–
–Kimi K2.7 Code––$4.0044

Pricing

  • Official API list price: $0.60 input and $3.00 output per 1M tokens (USD).
  • At a typical 3:1 input-to-output ratio, the blended cost is about $1.20 per 1M tokens.
  • Among 79 models with an output price, it is the #23 cheapest.
  • The cheapest OpenRouter route charges $0.53 input / $3.33 output.

Sources & methodology

Kimi K2.5 is not on the arena.ai Agents leaderboard, so it has no rank on our default leaderboard; the other data below is for reference.

  • arena.ai Agents — agent Net Improvement, 95% interval, sessions and rank (via the public Hugging Face dataset, CC BY 4.0).
  • LMArena — Arena score (Text leaderboard with style control, the same view arena.ai shows by default), 95% confidence interval, votes and rank.
  • Artificial Analysis — Intelligence / Coding Index, output speed and latency (medians).
  • OpenRouter — Route prices, context length and listing date.

Arena scores use the Bradley-Terry model (not Elo); ± is the half-width of the 95% confidence interval. Latency figures are Artificial Analysis medians; for reasoning models, time to first token and to first answer include the thinking phase. Data syncs daily at 14:20 (UTC+8).

Limits: Arena scores come from anonymous users voting on chats, so they reflect conversational preference rather than agent or coding benchmarks; with few votes the interval is wide and ranks move. Prices and speeds change as providers update them.

FAQ

What rank is Kimi K2.5 right now?

Kimi K2.5 is #54 on our leaderboard of 70 scored models (Arena score 1449.5); its raw LMArena rank is #81.

What does Kimi K2.5 cost via API?

Input $0.60 / output $3.00 per 1M tokens (official API list price; the OpenRouter quote when no official price is available). At a typical 3:1 input-to-output ratio, the blended cost is about $1.20 per 1M tokens.

How close is Kimi K2.5 to the models ranked next to it?

It is 3.7 points behind GPT-5.6 Luna (1453.2) and 4.0 points ahead of Gemini 2.5 Pro (1445.5); gaps smaller than the confidence interval (±3.3) are not significant.

Is Kimi K2.5 open source? Can it be self-hosted?

Kimi K2.5 is an open-weights model (license: Modified MIT); you can download the weights and self-host.

What is the context length of Kimi K2.5?

262K (262,144 tokens, as listed on OpenRouter).

Kimi K2.5 vs GPT-5.6 Luna: which is better?

On the LMArena Text leaderboard Kimi K2.5 scores 1449.5 and GPT-5.6 Luna scores 1453.2 (a 3.7-point gap, within the confidence interval, so not significant). On price, Kimi K2.5 costs $3.00 and GPT-5.6 Luna $1.20 per 1M output tokens.

Is there a cheaper model close to Kimi K2.5?

Yes: Qwen3.7 Plus scores 1455.3 on the Text leaderboard (Kimi K2.5: 1449.5) and costs $1.60 per 1M output tokens, about 47% less than Kimi K2.5 ($3.00).

Where does the Kimi K2.5 data come from and how often is it updated?

Agent success rates come from the arena.ai Agents leaderboard (CC BY 4.0), Text scores from LMArena, intelligence and speed from Artificial Analysis, and prices and context from official list prices and OpenRouter. 17NAS syncs daily at 06:20 UTC; this page was last updated Nov 10, 2026.

Related leaderboards