MiMo-V2.5

Language Xiaomi Verified Retired Shut down Oct 21, 2026 Replaced by MiMo-V2.6-Flash Released Apr 23, 2026
  • In Input modalities: TextImageVideoAudio Out Output modalities: Text
  • Reasoning
  • Reasoning can be off
  • Tool use
  • Structured output
  • Open weights

Prices

2 offerings · prices as listed
Xiaomi MiMo API (opens in a new tab) Official Retired China · CNY mimo-v2.5
Conditions/ 1M tokensInputCache readOutput
Standard no longer listed since Oct 28, 2026 ¥1.00¥0.020¥2.00

Prices from Xiaomi MiMo: pay-as-you-go pricing (opens in a new tab) (checked Oct 19, 2026)

Xiaomi MiMo API (opens in a new tab) Official Retired Global · USD mimo-v2.5
Conditions/ 1M tokensInputCache readOutput
Standard no longer listed since Oct 28, 2026 $0.14$0.0028$0.28

Prices from Xiaomi MiMo: pay-as-you-go pricing (opens in a new tab) (checked Oct 19, 2026)

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

Details

Knowledge cutoff
Dec 2024
Licence
mit
Family
mimo
Model ID

Not stated on Xiaomi’s pages: max input, reasoning efforts, max thinking budget, built-in tools, aliases.

Run locally

Weights: XiaomiMiMo/MiMo-V2.5
  • Parameters 310.8B
  • Active 15B · 256 experts, 8 per token
  • Native precision BF16
  • Layers 48
  • KV cache per token 144 KB
  • Config positions 1024K (config.json max_position_embeddings)
  • API context 977K (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
≈ 162.6 GB (weights 153.8 + KV cache 1.1 + overhead 7.7)
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 file293.4 GB≈ 309.2Doesn’t fit—
Q8_0307.5 GB≈ 324.0Doesn’t fit—
FP8289.4 GB≈ 305.0Doesn’t fit—
Q6_K237.7 GB≈ 250.7Doesn’t fit—
Q5_K_M205.8 GB≈ 217.3Doesn’t fit—
Q5_0199.0 GB≈ 210.0Doesn’t fit—
Q4_K_M175.8 GB≈ 185.7Doesn’t fit—
Q4_0162.8 GB≈ 172.1Doesn’t fit—
IQ4_XS153.8 GB≈ 162.6Doesn’t fit—
Q3_K_Mquality drops141.5 GB≈ 149.6Doesn’t fit—
IQ3_XXSquality drops110.7 GB≈ 117.4Doesn’t fit—
Q2_Kquality drops108.5 GB≈ 115.1Doesn’t fit—
IQ2_XXSquality drops74.5 GB≈ 79.4Doesn’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 GPU; unified memory Apple M5 Ultra (64-core GPU) · 256 GB.

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≈ 4632.1 GBDoesn’t fit your hardware
LoRA≈ 580.8 GBDoesn’t fit your hardware
QLoRA≈ 164.8 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 2026-11-06. How it’s estimated

Cited scores

From third-party leaderboards, not measured by us
  • LMArena Text 1,434

    #71 of 163 listed models · CI 1,430–1,438 · 47,327 votes

  • LMArena Text · Chinese 1,482

    #60 of 161 listed models · CI 1,469–1,494 · 2,586 votes

  • LMArena Text · Coding 1,491

    #64 of 163 listed models · CI 1,484–1,497 · 13,659 votes

  • LMArena Text · Creative writing 1,394

    #74 of 163 listed models · CI 1,387–1,402 · 7,876 votes

  • LMArena (Arena), leaderboard-dataset on Hugging Face, licensed CC BY 4.0 · License · as of Oct 19 – 28, 2026
  • changes we made
  1. Available → Retired on Xiaomi MiMo API
  2. Available → Retired on Xiaomi MiMo API, China
  3. Deprecated → Retired
  4. New language model

Sources

Xiaomi · China · mimo.xiaomi.com

Snapshots, the same family and other models from this provider