Arao AI

NVIDIA RTX 4090 D

GPU

Ada Lovelace; 24GB VRAM, 1008 GB/s bandwidth, 425W TDP, CUDA cores 14,592. Release date: 2023-12-28. Source: LCZ hardware catalog, with missing fields supplemented from NVIDIA official materials.

16

Run records

1

Models covered

4

Frameworks covered

L2 Cross-framework verified × 16

Evidence mix

Typical measured performance

Published benchmark averages per model × framework × quantization on this hardware.

  1. llama.cpp · IQ4_XS Decode 103 · Prefill 1079 ·
  2. llama.cpp · Q4_K_M Decode 92 · Prefill 992 ·
  3. llama.cpp · Q5_K_M Decode 82 · Prefill 977 ·
  4. llama.cpp · Q8_0 Decode 69 · Prefill 646 ·
  5. vLLM · IQ4_XS Decode 78 · Prefill 936 ·
  6. vLLM · Q4_K_M Decode 72 · Prefill 862 ·
  7. vLLM · Q5_K_M Decode 79 · Prefill 770 ·
  8. vLLM · Q8_0 Decode 57 · Prefill 577 ·
  9. Ollama · IQ4_XS Decode 101 · Prefill 1055 ·
  10. Ollama · Q4_K_M Decode 92 · Prefill 930 ·
  11. Ollama · Q5_K_M Decode 87 · Prefill 850 ·
  12. Ollama · Q8_0 Decode 54 · Prefill 647 ·
  13. Strata · IQ4_XS Decode 151 · Prefill 1357 ·
  14. Strata · Q4_K_M Decode 136 · Prefill 1276 ·
  15. Strata · Q5_K_M Decode 110 · Prefill 1311 ·
  16. Strata · Q8_0 Decode 93 · Prefill 1001 ·

Key specs

Vendor
NVIDIA
Architecture
Ada Lovelace
VRAM
24 GB
Memory bandwidth
1008 GB/s
TDP
425 W
Release date
2023-12-28

Run records

16 configurations

ModelFrameworkQuantizationDecodeGen timeSamples
Qwen3.5-27Bllama.cppIQ4_XS103 tok/s—1
Qwen3.5-27Bllama.cppQ4_K_M92 tok/s—1
Qwen3.5-27Bllama.cppQ5_K_M82 tok/s—1
Qwen3.5-27Bllama.cppQ8_069 tok/s—1
Qwen3.5-27BvLLMIQ4_XS78 tok/s—1
Qwen3.5-27BvLLMQ4_K_M72 tok/s—1
Qwen3.5-27BvLLMQ5_K_M79 tok/s—1
Qwen3.5-27BvLLMQ8_057 tok/s—1
Qwen3.5-27BOllamaIQ4_XS101 tok/s—1
Qwen3.5-27BOllamaQ4_K_M92 tok/s—1
Qwen3.5-27BOllamaQ5_K_M87 tok/s—1
Qwen3.5-27BOllamaQ8_054 tok/s—1
Qwen3.5-27BStrataIQ4_XS151 tok/s—1
Qwen3.5-27BStrataQ4_K_M136 tok/s—1
Qwen3.5-27BStrataQ5_K_M110 tok/s—1
Qwen3.5-27BStrataQ8_093 tok/s—1