GPU AFaster
NVIDIA RTX 5090
- VRAM
- 32 GB
- Bandwidth
- 1792 GB/s
- Street price
- $3,699
- Vendor
- nvidia
GPU B
NVIDIA RTX 4090
- VRAM
- 24 GB
- Bandwidth
- 1008 GB/s
- Street price
- $1,999
- Vendor
- nvidia
The short answer
Both GPUs handle the same models, but NVIDIA RTX 5090 is more than 20% faster on 31 of them. For the remaining 2 models, speeds are within 20% — you won't notice the gap. NVIDIA RTX 5090 gives consistently better throughput.
31 models A is 20%+ faster
2 equal (both run, <20% diff)
21 too large for either
Model-by-model fit
Click any row for the full breakdown. Tie% shown in the Winner column when both GPUs run the model within 20% of each other.
| Model | NVIDIA RTX 5090 | NVIDIA RTX 4090 | Winner |
|---|---|---|---|
c4ai-command-r-v01 35B 35B · command | 53 tok/s · Q6_K | 42 tok/s · Q4_K_M | A (faster) |
Command-R+ 104B 104B · command | Too large | Too large | — |
DeepSeek R1 Distill Llama 8B 8B · deepseek | 92 tok/s · FP16 | 52 tok/s · FP16 | A (faster) |
DeepSeek R1 Distill Qwen 14B 14.8B · deepseek | 51 tok/s · FP16 | 56 tok/s · Q8_0 | ≈tie B +10% |
DeepSeek R1 Distill Llama 70B 70.6B · deepseek | Too large | Too large | — |
DeepSeek R1 671B 671B · deepseek | Too large | Too large | — |
DeepSeek-V3 685B 685B · deepseek | Too large | Too large | — |
DeepSeek-V3.2 685.4B 685.4B · deepseek | Too large | Too large | — |
gemma-2-9b 9.2B · gemma | 81 tok/s · FP16 | 45 tok/s · FP16 | A (faster) |
gemma-2-27b 27.2B · gemma | 55 tok/s · Q8_0 | 38 tok/s · Q6_K | A (faster) |
Llama 3.1 8B Compact 8B · llama | 83 tok/s · FP16 | 47 tok/s · FP16 | A (faster) |
CodeLlama 34B 34B · llama | 54 tok/s · Q6_K | 44 tok/s · Q4_K_M | A (faster) |
CodeLlama 34B 34B · llama | 54 tok/s · Q6_K | 44 tok/s · Q4_K_M | A (faster) |
Llama 3.3 70B 70.6B · llama | Too large | Too large | — |
Llama 3.1 70B 70.6B · llama | Too large | Too large | — |
Llama 4 Scout 17B 109B · llama | Too large | Too large | — |
Llama-4-Maverick-17B-128E 400B · llama | Too large | Too large | — |
Llama 3.1 405B 405B · llama | Too large | Too large | — |
Mistral 7B v0.1 7.25B · mistral | 102 tok/s · FP16 | 57 tok/s · FP16 | A (faster) |
Codestral 22B 22.2B · mistral | 66 tok/s · Q8_0 | 46 tok/s · Q6_K | A (faster) |
Mixtral 8x7B Instruct v0.1 47B · mixtral | Too large | Too large | — |
Mistral Large 2 123B 123B · mistral | Too large | Too large | — |
Phi-4-mini 3.8B 3.8B · phi | 188 tok/s · FP16 | 106 tok/s · FP16 | A (faster) |
Phi-4 14B 14B · phi | 95 tok/s · Q8_0 | 53 tok/s · Q8_0 | A (faster) |
Qwen 2.5 1.5B 1.5B · qwen | 437 tok/s · FP16 | 246 tok/s · FP16 | A (faster) |
Qwen 2.5 3B 3.1B · qwen | 228 tok/s · FP16 | 128 tok/s · FP16 | A (faster) |
Qwen3.5-4B 4.7B · qwen | 154 tok/s · FP16 | 87 tok/s · FP16 | A (faster) |
Qwen 2.5 7B 7.6B · qwen | 97 tok/s · FP16 | 55 tok/s · FP16 | A (faster) |
Qwen 2.5 7B 7.6B · qwen | 97 tok/s · FP16 | 55 tok/s · FP16 | A (faster) |
Qwen 3 8B 8B · qwen | 83 tok/s · FP16 | 47 tok/s · FP16 | A (faster) |
Qwen3.5-9B 9.7B · qwen | 77 tok/s · FP16 | 43 tok/s · FP16 | A (faster) |
Qwen 3 32B 32B · qwen | 51 tok/s · Q6_K | 41 tok/s · Q4_K_M | A (faster) |
Qwen3.5-35B-A3B 36B · qwen | 51 tok/s · Q6_K | 41 tok/s · Q4_K_M | A (faster) |
Qwen 2.5 72B 72.7B · qwen | Too large | Too large | — |
Qwen 2.5 72B 72.7B · qwen | Too large | Too large | — |
Llama 3.2 1B 1.24B · llama | 534 tok/s · FP16 | 300 tok/s · FP16 | A (faster) |
Llama 4 Scout 17B 109B · llama | Too large | Too large | — |
DeepSeek R1 671B 671B · deepseek | Too large | Too large | — |
Gemma 3 27B 27B · gemma | 60 tok/s · Q6_K | 40 tok/s · Q5_K_M | A (faster) |
Qwen 3 8B 8B · qwen | 83 tok/s · FP16 | 47 tok/s · FP16 | A (faster) |
Qwen 3 32B 32B · qwen | 51 tok/s · Q6_K | 41 tok/s · Q4_K_M | A (faster) |
Llama 3.1 8B Compact 8B · llama | 83 tok/s · FP16 | 47 tok/s · FP16 | A (faster) |
Mixtral 8x7B Instruct v0.1 47B · mixtral | Too large | Too large | — |
Mistral Small 3.2 24B 24B · mistral | 65 tok/s · Q8_0 | 45 tok/s · Q6_K | A (faster) |
Command A 111B 111B · command | Too large | Too large | — |
DeepSeek R1 0528 685B · deepseek | Too large | Too large | — |
DeepSeek-V3-0324 684.5B · deepseek | Too large | Too large | — |
DeepSeek-R1-0528-Qwen3-8B 8.2B · qwen | 95 tok/s · FP16 | 53 tok/s · FP16 | A (faster) |
Qwen3-235B-A22B-Instruct-2507 235B · qwen | Too large | Too large | — |
Qwen3-30B-A3B-Instruct-2507 30B · qwen | 52 tok/s · Q8_0 | 52 tok/s · Q4_K_M | equal |
Qwen3-4B-Instruct-2507 4B · qwen | 195 tok/s · FP16 | 110 tok/s · FP16 | A (faster) |
gemma-4-E4B-it 8B · gemma | 97 tok/s · FP16 | 55 tok/s · FP16 | A (faster) |
gemma-4-26B-A4B-it 26.5B · gemma | 59 tok/s · Q8_0 | 41 tok/s · Q6_K | A (faster) |
gemma-4-31B-it 32.7B · gemma | 59 tok/s · Q6_K | 48 tok/s · Q4_K_M | A (faster) |
Want a different pairing? Browse all comparisons →