Kimi K2.5 vs Kimi K3

Moonshot AI · China  |  Moonshot AI · China · Updated June 2026

Quick verdict

Both are Moonshot AI models. Kimi K3 is the newer, generally stronger default; reach for Kimi K2.5 when its lower price or a specific cost or latency profile matters more than the latest capabilities.

Kimi K2.5 and Kimi K3 are both Moonshot AI models, so the real question is not which lab to trust but which tier fits your workload and budget. Kimi K2.5 is moonshot's open-weight multimodal model at $0.60/$2.50 with a 256K window — strong visual coding and agentic work, since superseded by K2.6/K2.7. Kimi K3 is moonshot's 2.8-trillion-parameter open-weight model — the largest open AI at release, with a 1M context and strong self-reported coding scores. Since both come from the same lab, the comparison below focuses on the tier-and-cost trade-offs that actually separate them.

Key differences at a glance

Side-by-side specs

SpecKimi K2.5Kimi K3
ProviderMoonshot AI (China) Moonshot AI (China)
ReleasedJanuary 27, 2026 July 27, 2026
Context window256K (~393 pages) 1M (~1,573 pages)
Price (in/out)$0.6/$2.5 per 1M tokens $3/$15 per 1M tokens
Open weight?Yes — self-hostable Yes — self-hostable
Modalitiestext, image, code text, image, video, code
SWE-Bench VerifiedNot published Not published
MRCR v2 @ 1MNot published Not published

Who wins what

Native multimodal reasoning and visual coding

Kimi K2.5

Kimi K3 is comparatively weak here — coding scores use Moonshot FrontierSWE, not standard SWE-Bench Verified

Agentic tool-calling and self-directed multi-step work

Kimi K2.5

Moonshot's open-weight multimodal model at $0.60/$2.50 with a 256K window — strong visual coding and agentic work, since superseded by K2.6/K2.7 — and it runs cheaper at $0.6/$2.5 per 1M tokens.

Open-weight (Modified-MIT) — self-hostable at 256K context

Kimi K2.5

Kimi K3 is comparatively weak here — 2.8T params need serious hardware to self-host — weights are free, running is not

Largest open-weight model at release — 2.8T sparse MoE, self-hostable

Kimi K3

Moonshot's 2.8-trillion-parameter open-weight model — the largest open AI at release, with a 1M context and strong self-reported coding scores — and it carries the larger 1M context.

1M-token context with native vision (text, image and video)

Kimi K3

Its 1M window holds about 4× more than Kimi K2.5's 256K in a single prompt.

Vendor reports 81.2 FrontierSWE and 88.3 Terminal-Bench 2.1 on its own harness

Kimi K3

Kimi K2.5 is comparatively weak here — its coding score uses Moonshot's own harness, not the standard SWE-Bench Verified protocol

Lowest cost at scale

Kimi K2.5

At $0.6/$2.5 per 1M tokens, it is the cheaper of the two — the gap dominates the bill on high-volume workloads.

Largest single-prompt input

Kimi K3

Its 1M window is about 4× larger than Kimi K2.5's 256K, fitting roughly 1,573 pages in one prompt.

Which should you pick?

A cost-sensitive startup shipping high volume

Kimi K2.5

At $0.6/$2.5 per 1M tokens it undercuts Kimi K3, and on millions of tokens that margin decides the monthly bill.

Someone analysing very long documents or codebases

Kimi K3

Larger 1M window fits more in one prompt.

Anyone whose priority is native multimodal reasoning and visual coding

Kimi K2.5

It is specifically built for that.

Anyone whose priority is largest open-weight model at release — 2.8t sparse moe, self-hostable

Kimi K3

That is its strongest area.

Kimi K2.5: where it fits

Moonshot's open-weight multimodal model at $0.60/$2.50 with a 256K window — strong visual coding and agentic work, since superseded by K2.6/K2.7. Released January 27, 2026 by Moonshot AI, it is built for native multimodal reasoning and visual coding, agentic tool-calling and self-directed multi-step work, open-weight (Modified-MIT) — self-hostable at 256K context, and vendor reports around 76.8% on its own SWE-agent coding harness.

Its trade-offs are real: its coding score uses Moonshot's own harness, not the standard SWE-Bench Verified protocol, superseded within Moonshot's line by Kimi K2.6 and K2.7, openRouter shows a promo price below Moonshot's $0.60/$2.50 list, and image input but no audio or video. At $0.6 in / $2.5 out per million tokens, it sits in the budget price band.

Kimi K3: where it fits

Moonshot's 2.8-trillion-parameter open-weight model — the largest open AI at release, with a 1M context and strong self-reported coding scores. Released July 27, 2026 by Moonshot AI, it is built for largest open-weight model at release — 2.8T sparse MoE, self-hostable, 1M-token context with native vision (text, image and video), vendor reports 81.2 FrontierSWE and 88.3 Terminal-Bench 2.1 on its own harness, and fresh-input pricing of $3/M (cached $0.30/M), flat across the full 1M context.

Its trade-offs: coding scores use Moonshot FrontierSWE, not standard SWE-Bench Verified, 2.8T params need serious hardware to self-host — weights are free, running is not, no independent benchmark reproduction yet at release, and image input but no audio or video. At $3 in / $15 out per million tokens, it sits in the mid price band.

The bottom line for this matchup

Because Kimi K2.5 and Kimi K3 come from the same lab (Moonshot AI), they share the same training philosophy and ecosystem — the decision is purely tier vs. cost. Kimi K3 is the more capable, more recent option; the other earns its place only when its price or latency profile fits a specific job better. Most teams should default to Kimi K3 and drop down only with a concrete reason.

Want both Kimi K2.5 and Kimi K3 without two subscriptions? LumiChats gives you these plus 40+ models under one ₹69/day pass (about $1/day) — draft with one, cross-check with the other.

See pricing

Frequently asked questions

Is Kimi K2.5 or Kimi K3 better for coding?

Public SWE-Bench figures are not available for either model, so the honest test is your own repository — run an identical real bug through both. By design, Kimi K2.5 leans toward native multimodal reasoning and visual coding while Kimi K3 leans toward largest open-weight model at release — 2.8t sparse moe, self-hostable, and that positioning usually predicts which feels better on your codebase.

Which is cheaper, Kimi K2.5 or Kimi K3?

Kimi K2.5 is cheaper — $0.6/$2.5 per 1M tokens vs $3/$15 per 1M tokens, roughly 5× apart on input.

Which has the bigger context window?

Kimi K3 — 1M vs 256K, about 4× larger. Useful only if the model actually reasons over the full window, which not all do.

Should I upgrade from Kimi K2.5 to Kimi K3?

Since both are Moonshot AI models, the newer one (Kimi K3) is usually the better default unless you need a specific cost or latency profile from the other.

Which is newer, Kimi K2.5 or Kimi K3?

Kimi K3 — released July 27, 2026, about 6 months after Kimi K2.5.

Related comparisons

Specifications and benchmarks reflect publicly reported figures as of June 2026 and may change as providers release updates. Always verify on your own workload.