Configs › Gemma-4-31B-IT › Gemma-4-31B-IT on DGX Spark: 11 tok/s, 3 measured configs
Gemma-4-31B-IT on DGX Spark: 11 tok/s, 3 measured configs
Measured Gemma-4-31B-IT speeds on DGX Spark: fastest 11 tok/s single-stream (Q4_K_M, llama.cpp/cuda), 3 configs with quant, backend and context, each linked to its source, snapshot 2026-10-29.
Snapshot 2026-10-29 · 413 configs on AI boxes and 288 on GPU cards · 121 models. Every number is generated from the tracker snapshot and linked to where it was measured; nothing here is typed by hand.
Fastest measured single-stream run of Gemma-4-31B-IT on DGX Spark: 11 tok/s (Q4_K_M, llama.cpp/cuda), snapshot 2026-10-29.
Every measured single-stream run of Gemma-4-31B-IT on DGX Spark
| Decode tok/s | Quant | Backend | Context | Trust | Measured by | Source checked | Source |
|---|---|---|---|---|---|---|---|
| 11 | Q4_K_M | llama.cpp/cuda | Structured table | shamily | 2026-10-26 | source | |
| 10.7 | Q4_K_M | llama.cpp/cuda | Structured table | nabe2030 | 2026-10-26 | source | |
| 6.6 | NVFP4 | vllm | Structured table | jvr0x | 2026-10-29 | source |
Best quant on DGX Spark
Fastest plain-decoding run per quant, versus the fastest quant.
- Q4_K_M: 11 tok/s, fastest (llama.cpp/cuda)
- NVFP4: 6.6 tok/s, 40% lower (vllm)
Backends on DGX Spark
Fastest plain-decoding run per backend. Quants can differ between rows.
- llama.cpp/cuda: 11 tok/s (Q4_K_M)
- vllm: 6.6 tok/s (NVFP4)
Gemma-4-31B-IT on other boxes
- Mac Ultra: 13.6 tok/s best of 1 config
Questions and answers
- How fast is Gemma-4-31B-IT on DGX Spark?
- The fastest of 3 measured configs is 11 tok/s single-stream (Q4_K_M, llama.cpp/cuda). The slowest is 6.6 tok/s.
- Which quant of Gemma-4-31B-IT is fastest on DGX Spark?
- Q4_K_M at 11 tok/s, ahead of NVFP4 at 6.6. Plain decoding only.
- Which backend is fastest for Gemma-4-31B-IT on DGX Spark?
- llama.cpp/cuda at 11 tok/s, then vllm at 6.6. Each backend figure is its best plain-decoding row and may use a different quant.
- How does DGX Spark compare with other boxes for Gemma-4-31B-IT?
- DGX Spark best 11 tok/s. Other boxes, best measured: Mac Ultra 13.6. Backends and quants differ, so this is the best known number per box, not a controlled test.
See also: Gemma-4-31B-IT on every box · DGX Spark hardware page (from S$7,452 in Singapore today).
Last updated: snapshot 2026-10-29; newest source check 2026-10-29.
More: all configs · hardware · news · contributors · methodology · llms.txt · llms-full.txt · API (OpenAPI) · JSON snapshot. Built by Altronis, private on-prem AI, Singapore.