Weekly archive

New local AI models, Apr 21–27, 2025

13 open models released Apr 21–27, 2025, with how each runs on five reference devices.

  1. Qwen3 235B-A22B

    Alibaba · LLM · 235B (22.0B active) ·

    83Great quality (A) · 83 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  2. Qwen3 32B

    Alibaba · LLM · 33B ·

    72Great quality (A) · 72 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  3. Qwen3 30B-A3B

    Alibaba · LLM · 31B (3.3B active) ·

    74Great quality (A) · 74 of 100: how capable the model is at F16, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  4. Cosmos-Predict2 14B Video2World

    NVIDIA · Video · 14.0B ·

    70Great quality (A) · 70 of 100: how capable the model is at FP16, whatever it runs on. Match score 62/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  5. Qwen3 14B

    Alibaba · LLM · 15B ·

    67Good quality (B) · 67 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  6. Llama Guard 4 12B

    Meta · Decision · 12B ·

    65Good quality (B) · 65 of 100: how capable the model is at Q6_K, whatever it runs on. Match score 73/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  7. Qwen3 8B

    Alibaba · LLM · 8.2B ·

    60Good quality (B) · 60 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 69/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  8. Cosmos-Predict2 2B Video2World

    NVIDIA · Video · 2.0B ·

    58Good quality (B) · 58 of 100: how capable the model is at FP16, whatever it runs on. Match score 59/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  9. Seed-Coder 8B Reasoning

    ByteDance · LLM · 8.3B ·

    58Good quality (B) · 58 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 68/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  10. Seed-Coder 8B Instruct

    ByteDance · LLM · 8.3B ·

    56Good quality (B) · 56 of 100: how capable the model is at Q8_0, whatever it runs on. Match score 67/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  11. Qwen3 4B

    Alibaba · LLM · 4.0B ·

    50Fair quality (C) · 50 of 100: how capable the model is at F16, whatever it runs on. Match score 64/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  12. Qwen3 1.7B

    Alibaba · LLM · 2.0B ·

    30Basic quality (D) · 30 of 100: how capable the model is at F16, whatever it runs on. Match score 55/100 (quality, fit, speed and memory use) sets the ranking.

    Runs on

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

  13. Qwen3 0.6B

    Alibaba · LLM · 0.8B ·

    14Basic quality (D) · 14 of 100: how capable the model is at F16, whatever it runs on. Match score 48/100 (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?