Kimi K2.5

Language Moonshot AI Partly verified Released Jan 2026

Kimi K2.5 combines visual and textual processing for agent workflows. Built on Kimi-K2-Base through further training with visual and text data, it is available as open source. Summary written by AI Index from Moonshot AI’s own pages.

  • In Input modalities: TextImageVideo Out Output modalities: Text
  • Reasoning
  • Reasoning can be off
  • Tool use
  • Open weights

Prices

9 offerings · prices as listed
Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 30, 2026 Global · USD Context: 262.14KReasoning: YesTool use: YesStructured output: —Reasoning can be off: YesMax output: 16.38KInput modalities: Text, Image, VideoOutput modalities: Text kimi-k2.5
Conditions/ 1M tokensInputCache readCache writeOutput
Standard $0.574$0.115—$3.011
Explicit cache —$0.057$0.718—

Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Nov 9, 2026) · kimi-k2.5 model (opens in a new tab) (checked Nov 8, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 30, 2026 China · CNY Context: 262.14KReasoning: YesTool use: YesStructured output: —Reasoning can be off: YesMax output: 16.38KInput modalities: Text, Image, VideoOutput modalities: Text kimi-k2.5
Conditions/ 1M tokensInputOutput
Standard ¥4.00¥21.00

Prices from 阿里云百炼:模型价格 (opens in a new tab) (checked Nov 9, 2026)

Moonshot AI Platform (opens in a new tab) Official Global · USD kimi-k2.5
Conditions/ 1M tokensInputCache readOutput
Standard past price, from the vendor’s page archived Mar 30, 2026$0.60$0.10$3.00

Prices from Moonshot AI platform pricing (opens in a new tab) (checked Oct 23, 2026)

Moonshot AI Platform (opens in a new tab) Official China · CNY kimi-k2.5
Conditions/ 1M tokensInputCache readOutput
Standard past price, from the vendor’s page archived Feb 17, 2026¥4.00¥0.70¥21.00

Prices from Kimi API 开放平台:模型推理价格 (opens in a new tab) (checked Oct 23, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 30, 2026 China · CNY kimi/kimi-k2.5
Conditions/ 1M tokensInputOutput
Standard ¥4.00¥21.00

Prices from 阿里云百炼:模型价格 (opens in a new tab) (checked Nov 9, 2026)

Alibaba Cloud Model Studio (opens in a new tab) Official Available on this channel since Jan 30, 2026 Germany (global scope) · US (global scope) · Japan (global scope) · USD Input modalities: Text, Image, VideoOutput modalities: TextTool use: YesContext: 262.14KMax output: 16.38KReasoning: YesStructured output: —Reasoning can be off: Yes kimi-k2.5
Conditions/ 1M tokensInputOutput
Standard no longer listed since Oct 23, 2026 $0.574$3.011

Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Nov 1, 2026)

Standard: the price a request with no conditions pays (the headline price elsewhere on the site).

Details

Max thinking budget
81.92K
Structured output
—
Knowledge cutoff
Jan 2025
Licence
Modified MIT License
Family
kimi-k
Model ID

Not stated on Moonshot AI’s pages: max input, reasoning efforts, built-in tools, aliases.

Run locally

Weights: moonshotai/Kimi-K2.5
  • Parameters 1.03T
  • Active 32B · 384 experts, 8 per token
  • Native precision BF16
  • Layers 61
  • KV cache per token 69 KB
  • Config positions 256K (config.json max_position_embeddings)
  • API context 250K (maker’s API)
My hardware GeForce RTX 4090 24 GB · 1008 GB/s · 64 GB DDR5 dual channel · 89.6 GB/s
On your hardware · GeForce RTX 4090
Doesn’t fit IQ4_XS (automatic) 8K context
WeightsKV cacheOverheadUsable
Needs
≈ 534.0 GB (weights 508.1 + KV cache 0.5 + overhead 25.4)
Usable
21.6 GB (90% of its memory)
Try
it’s too large for this hardware; see the smallest that runs it below
PrecisionWeightsNeeds at 8KOn your hardwareLongest context
BF16 (native)maker’s file554.3 GB≈ 582.6Doesn’t fit—
Q8_01016.1 GB≈ 1067.5Doesn’t fit—
FP8956.4 GB≈ 1004.7Doesn’t fit—
Q6_K785.4 GB≈ 825.2Doesn’t fit—
Q5_K_M680.2 GB≈ 714.8Doesn’t fit—
Q5_0657.5 GB≈ 690.9Doesn’t fit—
Q4_K_M581.0 GB≈ 610.6Doesn’t fit—
Q4_0538.0 GB≈ 565.4Doesn’t fit—
IQ4_XS508.1 GB≈ 534.0Doesn’t fit—
Q3_K_Mquality drops467.4 GB≈ 491.3Doesn’t fit—
IQ3_XXSquality drops365.8 GB≈ 384.6Doesn’t fit—
Q2_Kquality drops358.6 GB≈ 377.1Doesn’t fit—
IQ2_XXSquality drops246.3 GB≈ 259.1Doesn’t fit—

Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.

Compared with other hardware

Each at the highest precision it runs. Click one for details.

Smallest that runs it (Q4_K_M, 8K): no single listed machine; about 9 × 80 GB GPUs.

Run commands

The maker publishes no files at this precision. Choose its native precision or one of its own builds, or quantise it yourself with llama.cpp.

Memory to fine-tune
Full≈ 15305.6 GBDoesn’t fit your hardware
LoRA≈ 1917.4 GBDoesn’t fit your hardware
QLoRA≈ 542.7 GBDoesn’t fit your hardware

Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.

Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-21. How it’s estimated

Cited scores

From third-party leaderboards, not measured by us
  • Epoch Capabilities Index 148.7

    #64 of 123 listed models · CI 147.2–150.3

  • Epoch AI · SimpleQA Verified 34.3%

    #56 of 84 listed models · ±1.5

  • Epoch AI · SWE-bench Verified 73.8%

    #15 of 30 listed models · ±2

  • EQ-Bench Creative Writing v3 1,579

    #43 of 95 listed models

  • Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Oct 7 – Nov 14, 2026
  • EQ-Bench by Samuel J. Paech (eqbench.com; EQ-bench/EQ-bench-site on GitHub), MIT License. · License · as of Oct 28, 2026
  • changes we made

Included in coding plans

All coding plans →

  1. Now priced on Cloudflare Workers AI Input $0.60 · Cache read $0.10 · Output $3.00
  2. Now priced on Moonshot AI Platform, China Cache read ¥0.70 · Input ¥4.00 · Output ¥21.00
  3. Now priced on Moonshot AI Platform Cache read $0.10 · Input $0.60 · Output $3.00
  4. Now priced on Moonshot AI Platform, China Cache read ¥0.70 · Input ¥4.00 · Output ¥21.00
  5. Now priced on Moonshot AI Platform Cache read $0.10 · Input $0.60 · Output $3.00
  6. Now priced on Alibaba Cloud Model Studio, China Input ¥4.00 · Output ¥21.00
  7. Context window 256K → 262.14K
  8. Now priced on Alibaba Cloud Model Studio Input $0.574 · Output $3.011

Sources

Moonshot AI’s official policies: data use, privacy, terms →

Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data

Moonshot AI · China · moonshot.ai

Snapshots, the same family and other models from this provider