Recipe
deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1
deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1MiaAI-Lab single-Spark DeepSeek V4 Flash 0731 EXL3 3.0bpw profile with SparkInfer, DSpark K5, native NVFP4 DS-MLA KV, and graph captures
Record
- Status
- candidate
- Source
- mialabs
- Engine
- sparkinfer
- Engine version
- NVIDIA vLLM 26.02 base
- Graph
- full-and-piecewise
- Accelerators
- 1
- Tensor parallel
- 1
- Context tokens
- 370,104
- Max concurrency
- 1
- chat
- yes
- reasoning
- yes
- tools
- yes
- vision
- no
Hugging Face model card
Identity
https://huggingface.co/0xSero/deepseek-v4-flash-0731-spark- Repository
- 0xSero/deepseek-v4-flash-0731-spark
- Status
- known
- Link type
- Exact Hub repository
Public Hugging Face repository confirmed by the Hub API.
Candidate: useful compatibility or speed evidence. The registry does not offer Run until promotion requirements are met.
Script configuration
sparkinfer
No container · candidate · script
Candidate evidence — not a Run contract
Environment
| Variable | Value |
|---|---|
CUDAGRAPH_CAPTURE_SIZES | 6,12,24 |
GPU_MEMORY_UTILIZATION | 0.94 |
KV_RECORD | stock432 |
MAX_CUDAGRAPH_CAPTURE_SIZE | 24 |
MAX_MODEL_LEN | 384000 |
MAX_NUM_BATCHED_TOKENS | 8224 |
MAX_NUM_SEQS | 1 |
MODEL_REVISION | 22f28d32b9b29b4352eaa380ff8c2c170b2847ab |
Measured speed
| Concurrency | Context | Prefill | Decode | TTFT ms | Status | Sweep |
|---|---|---|---|---|---|---|
| 1 | 320,000 | 630 | 47 | — | historical | deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1-sweep |
| 1 | 370,104 | 625 | 44 | — | historical | deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1-sweep |
Remaining fields
Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.
- hardware count
- 1
- hardware id
- dgx-spark-gb10-128gb
- id
- deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1
- model instance id
- 0xsero-deepseek-v4-flash-0731-spark--3-0-bpw-reap-k216
- recipe source
- mialabs
- schema version
- local-ai-registry/v1
- speed sweep ids
- deepseek-v4-flash-0731-exl3-3bpw-dgxspark-sparkinfer-tp1-sweep
- status
- candidate
capabilities
- chat
- Yes
- reasoning
- Yes
- tools
- Yes
- vision
- No
engine
- graph mode
- full-and-piecewise
- name
- sparkinfer
- version
- NVIDIA vLLM 26.02 base
serving
- max concurrency
- 1
- max context tokens
- 370,104
- tensor parallel
- 1
Provenance & metadata (3)
facts
- serving.kv cache tokens · provenance · captured at
- 2026-08-27T06:04:13.773Z
serving.kv cache tokens · reason kv-cache-capacity-not-published
serving.kv cache tokens · state unknown
metadata
provenance
- captured at
- 2026-08-27T06:04:13.773Z
sources
| captured at | kind | url |
|---|---|---|
| 2026-08-27T06:04:13.773Z | normalized-recipe | github.com/0xSero/local-ai-registry ↗ |