Performance ranking of models that run in 16 GB of VRAM (Epoch Capabilities Index (Epoch AI))

1. Rank Model Index (95% interval) Quantization that fits 16 GB
2. 1
   Qwen3.8-27B

   Qwen 27.3B Qwen 3.8 27B

   Index (95% interval) 152.7 150.7–154.9

   UD-Q4_K_S 17.1 GB

3. 2
   Qwen3.6-27B

   Qwen 26.9B Qwen3.6 27B

   Index (95% interval) 146.7 144.6–148.9

   IQ4_XS 17.2 GB

4. 3
   gemma-4-31B-it

   google 30.7B Gemma 4 31B IT

   Index (95% interval) 144.2 141.8–146.8

   IQ3_XS 16.8 GB

5. 4
   Qwen3.6-35B-A3B

   Qwen 34.7B Qwen 3.6 35B-A3B

   Index (95% interval) 144.1 142.1–145.9

   IQ3_XXS 17.0 GB

6. 5
   Qwen3.5-35B-A3B

   Qwen 34.7B Qwen3.5-35B-A3B

   Index (95% interval) 144.0 141.5–146.6

   IQ3_XXS 17.0 GB

7. 6
   gemma-4-26B-A4B-it

   google 25.2B Gemma 4 26B A4B

   Index (95% interval) 141.7 139.2–144.2

   Q4_0 16.3 GB

8. 7
   Qwen3.5-9B

   Qwen 9B Qwen3.5-9B

   Index (95% interval) 139.4 137.1–141.5

   UD-Q8_K_XL 14.3 GB

9. 8
   Qwen3-30B-A3B-Thinking-2507

   Qwen 30.5B Qwen3-30B-A3B-Thinking (Jul 2025)

   Index (95% interval) 138.4 135.7–140.9

   Q3_K_M 16.4 GB

10. 9
    Qwen3-32B

    Qwen 32.8B Qwen3-32B

    Index (95% interval) 137.0 134.5–139.4

    UD-IQ3_XXS 16.0 GB

11. 10
    QwQ-32B

    Qwen 32.8B QwQ-32B

    Index (95% interval) 136.5 133.6–139.3

    Q2_K 15.3 GB

12. 11
    Qwen3-14B

    Qwen 14.8B Qwen3-14B

    Index (95% interval) 136.2 133.5–138.7

    UD-Q6_K_XL 15.5 GB

13. 12
    Qwen3-30B-A3B-Instruct-2507

    Qwen 30.5B Qwen3-30B-A3B-Instruct (Jul 2025)

    Index (95% interval) 135.2 132.3–138.0

    Q3_K_XL 16.5 GB

14. 13
    gpt-oss-20b

    openai 20.9B gpt-oss-20b

    Index (95% interval) 135.0 132.4–137.5

    F16 15.0 GB

15. 14
    Qwen3-30B-A3B

    Qwen 30.5B Qwen3-30B-A3B

    Index (95% interval) 134.2 131.3–136.9

    Q3_K_M 16.4 GB

16. 15
    Qwen3-8B

    Qwen 8.2B Qwen3-8B

    Index (95% interval) 132.2 129.1–135.1

    UD-Q8_K_XL 12.9 GB

17. 16
    phi-4

    microsoft 14.7B Phi-4

    Index (95% interval) 129.1 126.1–131.9

    Q6_K 14.5 GB

18. 17
    gemma-3-27b-it

    google 27B Gemma 3 27B

    Index (95% interval) 127.7 124.6–130.5

    IQ4_XS 17.2 GB

19. 18
    Qwen2.5-Coder-7B

    QuantFactory 7.6B Qwen2.5-Coder (7B)

    Index (95% interval) 108.0 98.7–116.4

    Q8_0 9.42 GB

20. 19
    Mistral-7B-v0.1

    mistralai 7.2B Mistral 7B v0.1

    Index (95% interval) 103.3 95.7–109.8

    Q8_0 9.37 GB

Estimated Memory at 8K: Estimated with a formula. Actual usage depends on your setup and settings.