DiffusionGemma 26B A4B needs ~21.1 GB VRAM. Tesla P40 24GB has 24.0 GB. With Q3_K_S quantization, expect ~28 tok/s.
Operating mode
Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.
Current mode
Balanced
Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.
Select quantization to explore
0.2 GB over capacity — needs offload or smaller quantization
Fit status
Too heavy
Decode
17.5 tok/s
TTFT
11073 ms
Safe context
15K
Memory
24.2 GB / 24.0 GB
This setup is broadly balanced for this model.
Older PCIe generation
PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | A | Tight fit | 24.4 tok/s | 4326 ms | 15K |
| Coding | F | Too heavy | 17.5 tok/s | 11073 ms | 15K |
| Agentic Coding | F | Too heavy | 12.8 tok/s | 21948 ms | 15K |
| Reasoning | F | Too heavy | 17.5 tok/s | 13086 ms | 15K |
| RAG | F | Too heavy | 12.8 tok/s | 27434 ms | 15K |
Inference speed
Estimated decode speed (tokens/sec) for DiffusionGemma 26B A4B at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~144 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.
| GPU / Mac | Memory | Quant | Speed (tok/s) | Fits? |
|---|---|---|---|---|
 RTX 5090 32GB | 32 GB | Q4_K_M | 143.6 | Fits |
| RTX 4090 24GB | 24 GB | Q4_K_M | 67.5 | Too big |
Mac Studio M3 Ultra 256GB | 256 GB | Q4_K_M | 66.6 | Fits |
RX 7900 XTX 24GB | 24 GB | Q4_K_M | 60.9 | Too big |
| RTX 3090 24GB | 24 GB | Q4_K_M | 57.7 | Too big |
Mac Studio M2 Ultra 128GB | 128 GB | Q4_K_M | 55.5 | Fits |
Mac Studio M1 Ultra 128GB | 128 GB | Q4_K_M | 52.6 | Fits |
MacBook Pro M4 Max 128GB | 128 GB | Q4_K_M | 41.1 | Fits |
MacBook Pro M4 Max 64GB | 64 GB | Q4_K_M | 41.1 | Fits |
MacBook Pro M3 Max 64GB | 64 GB | Q4_K_M | 28.7 | Fits |
MacBook Pro M1 Max 64GB | 64 GB | Q4_K_M | 26.3 | Fits |
MacBook Pro M4 Pro 48GB | 48 GB | Q4_K_M | 25.1 | Fits |
| RTX 4080 Super 16GB | 16 GB | Q4_K_M | 24.6 | Too big |
| RTX 4070 12GB | 12 GB | Q4_K_M | 8.6 | Too big |
| RTX 3060 12GB | 12 GB | Q4_K_M | 5.4 | Too big |
| RTX 4060 8GB | 8 GB | Q4_K_M | 3.6 | Too big |
Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.
How DiffusionGemma 26B A4B (25.799999237060547B params) fits at each quantization level on Tesla P40 24GB (24.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 10.1 GB | Low | A78 |
Q3_K_S | 3 | 12.6 GB | Low | A79 |
NVFP4 | 4 | 14.4 GB | Medium | A79 |
Q4_K_M | 4 | 15.7 GB | Medium | A78 |
Q5_K_MBest for your GPU | 5 | 18.6 GB | High | A78 |
Q6_K | 6 | 21.2 GB | High | F0 |
Q8_0 | 8 | 27.6 GB | Very High | F0 |
F16 | 16 | 52.9 GB | Maximum | F0 |
Copy-paste commands to run DiffusionGemma 26B A4B on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "google/diffusiongemma-26B-A4B-it" \
--hf-file "diffusiongemma-26B-A4B-it-Q4_K_M.gguf" \
-c 4096 -ngl 99Upgrade options
RTX 5090 32GBBudget pickMakes the model fit on the accelerator instead of staying completely out of reach.
Raises estimated decode speed by about 721%.
~$1,999 MSRP
RTX PRO 4500 Blackwell 32GBBest valueMakes the model fit on the accelerator instead of staying completely out of reach.
Raises estimated decode speed by about 414%.
~$2,499 MSRP
RTX 5000 Ada 32GBNVIDIA upgradeMakes the model fit on the accelerator instead of staying completely out of reach.
Raises estimated decode speed by about 215%.
~$4,000 MSRP
Yes, Tesla P40 24GB can run DiffusionGemma 26B A4B at Q3_K_S quantization (Tight fit). The recommended Q4_K_M requires 24.2 GB which exceeds available memory, but at Q3_K_S it needs only 21.1 GB. Expected decode speed: 28.3 tok/s.
DiffusionGemma 26B A4B (25.799999237060547B parameters) requires approximately 24.2 GB at Q4_K_M quantization. On Tesla P40 24GB, it fits at Q3_K_S using 21.1 GB.
The recommended quantization is Q4_K_M, but on Tesla P40 24GB the best fitting quantization is Q3_K_S, which uses 21.1 GB.
On Tesla P40 24GB, DiffusionGemma 26B A4B achieves approximately 28.3 tokens per second decode speed with a time-to-first-token of 6850ms using Q3_K_S quantization.
For coding workloads, DiffusionGemma 26B A4B on Tesla P40 24GB receives a F grade with 17.5 tok/s and 15K context.
On Tesla P40 24GB, DiffusionGemma 26B A4B can safely use up to 29K tokens of context at Q3_K_S quantization. The model's official context limit is 262K, but available memory constrains the safe maximum.
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/diffusiongemma-26b-a4b-on-tesla-p40-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: