Weekly archive

New local AI models, Sep 21–27, 2026

6 open models released Sep 21–27, 2026, with how each runs on five reference devices.

  1. MiMo V2.6 Pro (1T-A42B)

    Xiaomi · LLM · 1,024B (42.0B active) ·

    68Good quality (B) · 68 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 61/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

  2. MiMo V2.6 Flash (310B-A15B)

    Xiaomi · LLM · 311B (15.0B active) ·

    93Excellent quality (A+) · 93 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 97/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

  3. Naive N0.5 Flash (309B-A15.5B)

    Naive AI · LLM · 309B (15.5B active) ·

    92Excellent quality (A+) · 92 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 96/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

  4. Intern-Decision 4B

    Shanghai AI Lab · Decision · 4.5B ·

    84Great quality (A) · 84 of 100: how capable the model is at F16, whatever it runs on. Match score 92/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

  5. GLiNER2.5 Decide

    Fastino · Decision · 490M ·

    80Great quality (A) · 80 of 100: how capable the model is at F16, whatever it runs on. Match score 91/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

  6. MiMo V2.6 Distill Qwen 9B

    Xiaomi · LLM · 9.4B ·

    78Great quality (A) · 78 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 77/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

    Memory, every quant and how to run it Check your own device

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.

Was this page helpful?