Local AI Wizard
Xiaomi / language model; inputs not listed

MiMo V2.6 Pro

EVIDENCE AT A GLANCE

Quality, size and local speed

LiveBench 2026-06-25 overall ↗Model card / primary source ↗
LiveBench · overall—/ 100

No comparable score in this release

Model-weight disk · Q4 estimate~637.5 GBFormula, not a published download size
Planning memory · Q4 estimate~773 GBIncludes planning allowance; not measured VRAM
Local speedNo exact measurementDo not infer speed from quality or memory
Quality · No score in the reviewed releaseMemory · Planning estimate based on documented model size; not measured hardware fitLocal speed · No exact local measurement

LiveBench 2026-06-25 measures the named benchmark variant, not a local Q4 run. Scores were collected 2026-09-26; the seven categories share the same 0–100 scale. This is not the Artificial Analysis Intelligence Index. Q4 size covers only the counted weights; multimodal auxiliary components may add files.

SAME BENCHMARK RELEASE

Quality versus estimated Q4 weight size

Full ranking ↗

Reviewed language checkpoints with a LiveBench score and a parameter-based size estimate. The highlighted point is this model when both values are available.

506070809010100100010000Qwen3.8 27B · LiveBench 75.3 · Q4 ~16.9 GBQwen3.6 27B · LiveBench 64.0 · Q4 ~16.9 GBSmaug Mini · LiveBench 76.9 · Q4 ~17.5 GBGLM-5.3 Flash · LiveBench 71.6 · Q4 ~200.0 GBKimi K2.6 · thinking · LiveBench 70.5 · Q4 ~625.0 GBDeepSeek V4 Flash · LiveBench 65.5 · Q4 ~177.5 GBDeepSeek V4 Pro · LiveBench 71.6 · Q4 ~1000.0 GBKimi K2.7 Code · LiveBench 68.4 · Q4 ~625.0 GBMiniMax M3 · LiveBench 67.3 · Q4 ~267.5 GBKimi K3 · LiveBench 79.2 · Q4 ~1750.0 GBInkling · xhigh · LiveBench 71.9 · Q4 ~609.4 GBQwen3.8 Max · LiveBench 78.5 · Q4 ~1500.0 GBSmaug Agentic · LiveBench 79.5 · Q4 ~1750.0 GBDeepSeek V4 Pro · 0813 · LiveBench 77.4 · Q4 ~1000.0 GBNemotron 3 Ultra 550B A55B · LiveBench 67.4 · Q4 ~343.8 GB
Smaug Agentic79.5~1750.0 GBKimi K379.2~1750.0 GBQwen3.8 Max78.5~1500.0 GBDeepSeek V4 Pro · 081377.4~1000.0 GBSmaug Mini76.9~17.5 GB
Show 10 more modelsQwen3.8 27B75.3~16.9 GBInkling · xhigh71.9~609.4 GBGLM-5.3 Flash71.6~200.0 GBDeepSeek V4 Pro71.6~1000.0 GBKimi K2.6 · thinking70.5~625.0 GBKimi K2.7 Code68.4~625.0 GBNemotron 3 Ultra 550B A55B67.4~343.8 GBMiniMax M367.3~267.5 GBDeepSeek V4 Flash65.5~177.5 GBQwen3.6 27B64.0~16.9 GB

Horizontal axis: estimated Q4 model-weight disk size in decimal GB (log scale). Vertical axis: LiveBench overall / 100. This formula is not the published checkpoint size, runtime memory or a speed test. Models without both inputs are left out of this chart, not scored as zero.

The MiMo V2.6 Pro checkpoint is published by Xiaomi for language and reasoning. Its weights are available under MIT; that alone does not establish that its code and all components are open source.

When to consider it

Compare MiMo V2.6 Pro with other language and reasoning checkpoints using its linked publisher card. No score for it appears in the reviewed LiveBench snapshot. This is quality evidence, not a local speed result.

Check before choosing

The generic Q4 planning estimate is about 773 GB from 1020B stated parameters. It is not a measured minimum; context, batch size, runtime and quantization change actual memory use.

What is documented

Weight license
MIT ↗
Task and modality
language and reasoning · language model; inputs not listed
Count and basis
1020B · manufacturer specification · 42B active
Context
1,048,576 tokens published for this model
Quality evidence
No comparable score in the reviewed snapshot
Runtime and precision
No local execution verified. Check the publisher's supported runtime and checkpoint precision; batch and context memory are not measured here.

Run locally: known and unknown

~773 GB generic Q4 planning memory. Formula: parameters × 0.625 bytes, plus 20% working allowance and 8 GB reserve. This is neither measured VRAM nor checkpoint download size.

Q4 planning estimate only. Check checkpoint format, runtime support, KV cache, context, batch size and offload before choosing hardware.

Evidence level: Generic estimate only; no execution reproduced by this site.

Before downloading

  1. Check the exact checkpoint and license linked here.
  2. Confirm a compatible runtime and precision or quantization for your platform.
  3. Allow for context, batch size and any extra media components beyond the weights.

Catalog reviewed: 2026-09-30. Benchmark release: 2026-06-25; snapshot collected: 2026-09-26.