readygpu

NVIDIA GeForce RTX 5090 vs NVIDIA Tesla P40

For running models locally, priced in India. At 32k context the first runs 29 models fully in memory, the second 26.

In short

The NVIDIA GeForce RTX 5090 has more memory for models: 32 GB against 24 GB. At 32k context the NVIDIA GeForce RTX 5090 runs 29 of the models we track fully in memory and the NVIDIA Tesla P40 26; only the NVIDIA GeForce RTX 5090 holds Kimi Linear 48B-A3B. On the models both run fully in memory, the NVIDIA GeForce RTX 5090 is about 5.2× as fast as the NVIDIA Tesla P40 (estimated, median across models).

Updated . Speeds are estimates; how they are worked out.

NVIDIA GeForce RTX 5090

Not priced here

Add to build
NVIDIA Tesla P40

Not priced here

Add to build

Specifications

SpecNVIDIA GeForce RTX 5090NVIDIA Tesla P40
GPUGeForce RTX 5090Tesla P40
Memory32 GB GDDR724 GB GDDR5
Memory bus512-bit384-bit
Memory bandwidth1,792 GB/s346 GB/s
FP16 compute209 TFLOPS11.8 TFLOPS
SoftwareCUDACUDA
Board power575 W250 W
PCIe generationGen 5Gen 3
PCIe lanesx16x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelNVIDIA GeForce RTX 5090NVIDIA Tesla P40
Qwen 3.8 27B35tok/sQ8_010tok/sUD-Q5_K_M
Qwen 3.6 27B36tok/sQ8_010tok/sQ5_K_M
Gemma 4 31B39tok/sQ6_K10tok/sQ4_K_M
Qwen 3.6 35B-A3B210tok/sUD-Q6_K50tok/sUD-Q4_K_M
Qwen 3.5 27B36tok/sQ8_010tok/sQ5_K_M
Gemma 4 26B-A4B132tok/sQ8_029tok/sUD-Q6_K
Qwen 3.5 35B-A3B212tok/sQ6_K50tok/sQ4_K_M
GLM 4.7 Flash114tok/sQ6_K24tok/sQ5_K_M
Devstral Small 2 24B37tok/sQ8_010tok/sQ5_K_M
Kimi Linear 48B-A3B311tok/sQ4_K_Moffload
gpt-oss 20B206tok/sMXFP440tok/sMXFP4
Qwen3 Coder 30B-A3B117tok/sQ6_K26tok/sQ4_K_M
Qwen3 32B35tok/sQ5_K_Moffload
Gemma 4 12B82tok/sQ8_016tok/sQ8_0
Qwen 3.5 9B104tok/sQ8_020tok/sQ8_0
Mistral Small 3.2 24B37tok/sQ8_010tok/sQ5_K_M
DeepSeek R1 Distill Qwen 32B35tok/sQ5_K_Moffload
Gemma 3 27B35tok/sQ8_010tok/sQ5_K_M
Ministral 3 14B56tok/sQ8_011tok/sQ8_0
Qwen3 14B53tok/sQ8_010tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B82tok/sQ8_016tok/sQ8_0
Qwen3 8B82tok/sQ8_016tok/sQ8_0
Phi-4 14B50tok/sQ8_010tok/sQ8_0
Qwen 3.5 4B198tok/sQ8_038tok/sQ8_0
Ministral 3 8B82tok/sQ8_016tok/sQ8_0
Granite 4.2 8B76tok/sQ8_015tok/sQ8_0
Gemma 4 E4B121tok/sQ8_023tok/sQ8_0
LFM2.5 8B-A1B249tok/sQ8_048tok/sQ8_0
Llama 3.1 8B102tok/sQ6_K20tok/sQ6_K
Qwen 3.5 122B-A10Boffloadoffload
Qwen3 Coder Nextoffloadoffload
Nemotron 3 Super 120B-A12Boffloadwon't run
Mistral Small 4 119Boffloadoffload
GLM 4.5 Airoffloadoffload
Qwen3 Next 80B-A3Boffloadoffload
Nemotron 3.5 Lightning 30B-A3Boffloadoffload
Llama 3.3 70Boffloadwon't run
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
gpt-oss 120Bwon't runwon't run

Questions

Which is better for local AI, the NVIDIA GeForce RTX 5090 or the NVIDIA Tesla P40?
Choose the NVIDIA GeForce RTX 5090 to run bigger models. Choose the NVIDIA GeForce RTX 5090 for speed.
Which can run bigger models, the NVIDIA GeForce RTX 5090 or the NVIDIA Tesla P40?
At 32k context the NVIDIA GeForce RTX 5090 runs 29 of the models we track fully in memory and the NVIDIA Tesla P40 26; only the NVIDIA GeForce RTX 5090 holds Kimi Linear 48B-A3B.
Which is faster for local AI, the NVIDIA GeForce RTX 5090 or the NVIDIA Tesla P40?
On the models both run fully in memory, the NVIDIA GeForce RTX 5090 is about 5.2× as fast as the NVIDIA Tesla P40 (estimated, median across models).

More comparisons