Recipe

qwen3-6-27b-nvfp4-dgx-spark-gb10-128gb-vllm-tp1-62k8tt93

qwen3-6-27b-nvfp4-dgx-spark-gb10-128gb-vllm-tp1-62k8tt93

Observed LocalMaxxing leaderboard run. Evidence for compatibility, not an executable launch contract.

Record

Status
candidate
Source
localmaxxing
Engine
vllm
Engine version
0.26.0
Accelerators
1
Tensor parallel
1
Context tokens
2,048
Max concurrency
4
chat
unknown
reasoning
unknown
tools
unknown
vision
unknown

Hugging Face model card

Identity

https://huggingface.co/rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm
Repository
rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm
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.

Observed configuration

vllm

Evidence only · candidate · reference

Candidate evidence — not a Run contract

Source
https://www.localmaxxing.com/en/runs/cmqw0fns0034zqr0162k8tt93

Observed source tokens

Mechanical split of the source command. Unverified against the engine CLI. The registry does not offer Run for this recipe.

  1. vllm
  2. serve
  3. rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm
  4. --port
  5. 8000
  6. --trust-remote-code
  7. --quantization
  8. compressed-tensors
  9. --max-model-len
  10. 262144
  11. --gpu-memory-utilization
  12. 0.60
  13. --max-num-seqs
  14. 4
  15. --max-num-batched-tokens
  16. 8192
  17. --reasoning-parser
  18. qwen3
  19. --enable-auto-tool-choice
  20. --tool-call-parser
  21. qwen3_coder
  22. --enable-prefix-caching
  23. --enable-chunked-prefill
  24. --kv-cache-dtype
  25. fp8
  26. --attention-backend
  27. flashinfer
  28. --speculative-config
  29. {method:mtp,num_speculative_tokens:1}
  30. --moe-backend
  31. marlin
FlagValue
--port8000
--quantizationcompressed-tensors
--max-model-len262144
--gpu-memory-utilization0.60
--max-num-seqs4
--max-num-batched-tokens8192
--reasoning-parserqwen3
--tool-call-parserqwen3_coder
--kv-cache-dtypefp8
--attention-backendflashinfer
--speculative-config{method:mtp,num_speculative_tokens:1}
--moe-backendmarlin

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
2,04818.9observedqwen3-6-27b-nvfp4-dgx-spark-gb10-128gb-vllm-tp1-62k8tt93-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
qwen3-6-27b-nvfp4-dgx-spark-gb10-128gb-vllm-tp1-62k8tt93
model instance id
rdtand-qwen3-6-27b-prismascout-blackwell-nvfp4-bf16-vllm--nvfp4
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
qwen3-6-27b-nvfp4-dgx-spark-gb10-128gb-vllm-tp1-62k8tt93-sweep
status
candidate

capabilities

draft launch

accelerator backend
nvidia
arguments
serve, --model, rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm, --tensor-parallel-size, 1, --host, 0.0.0.0, --port, 8000, --max-model-len, 2048
container port
8,000
entrypoint
vllm
host port
8,000
image
vllm/vllm-openai@sha256:0a51ea5b4ae2dc5d81890e5173f54203d2a3ae0cfffe51b8fd2afd4391bfd967
ipc
host
kind
docker
shm size
16g
synthesized · generated at
2026-08-31T22:12:17Z
synthesized · image provenance
deepseek-fp8-rtx-pro-6000-blackwell-96gb-vllm-tp1
synthesized · template
vllm-openai-v1

mounts

read onlytarget
No/root/.cache/huggingface

engine

name
vllm
version
0.26.0

serving

max concurrency
4
max context tokens
2,048
tensor parallel
1
Provenance & metadata (3)

facts

capabilities.chat · provenance · captured at
2026-08-30T09:26:09Z
capabilities.chat · reason capability-not-verified
capabilities.chat · state unknown
capabilities.reasoning · provenance · captured at
2026-08-30T09:26:09Z
capabilities.reasoning · reason capability-not-verified
capabilities.reasoning · state unknown
capabilities.tools · provenance · captured at
2026-08-30T09:26:09Z
capabilities.tools · reason capability-not-verified
capabilities.tools · state unknown
capabilities.vision · provenance · captured at
2026-08-30T09:26:09Z
capabilities.vision · reason capability-not-verified
capabilities.vision · state unknown
draft launch.entrypoint · provenance · captured at
2026-09-01T01:41:26Z
draft launch.entrypoint · reason linux-amd64-container-config-entrypoint
draft launch.entrypoint · state known
engine.graph mode · provenance · captured at
2026-08-30T09:26:09Z
engine.graph mode · reason runtime-detail-not-published
engine.graph mode · state unknown
engine.version · provenance · captured at
2026-08-31T23:38:26Z
engine.version · reason copied-from-synthesized-image-provenance
engine.version · state known
serving.kv cache tokens · provenance · captured at
2026-08-30T09:26:09Z
serving.kv cache tokens · reason kv-cache-capacity-not-published
serving.kv cache tokens · state unknown
serving.max concurrency · provenance · captured at
2026-08-31T23:03:15Z
serving.max concurrency · reason server-capacity-derived-from-source-evidence
serving.max concurrency · state known

metadata

localmaxxing · backend
cuda
localmaxxing · batch size
1
localmaxxing · hardware label
GB10 Grace Blackwell
localmaxxing · notes
PrismaSCOUT mixed BF16/NVFP4 (~5.5 bit avg). MTP×1 speculative decoding. marlin MoE backend. ~21 GB weights. DeltaNet layers give 4× smaller KV cache.
localmaxxing · observed command
vllm serve rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm --port 8000 --trust-remote-code --quantization compressed-tensors --max-model-len 262144 --gpu-memory-utilization 0.60 --max-num-seqs 4 --max-num-batched-tokens 8192 --reasoning-parser qwen3 --enable-auto-tool-choice --tool-call-parser qwen3_coder --enable-prefix-caching --enable-chunked-prefill --kv-cache-dtype fp8 --attention-backend flashinfer --speculative-config {method:mtp,num_speculative_tokens:1} --moe-backend marlin
localmaxxing · run id
cmqw0fns0034zqr0162k8tt93
localmaxxing · tokenized · arguments
vllm, serve, rdtand/Qwen3.6-27B-PrismaSCOUT-Blackwell-NVFP4-BF16-vllm, --port, 8000, --trust-remote-code, --quantization, compressed-tensors, --max-model-len, 262144, --gpu-memory-utilization, 0.60, --max-num-seqs, 4, --max-num-batched-tokens, 8192, --reasoning-parser, qwen3, --enable-auto-tool-choice, --tool-call-parser, qwen3_coder, --enable-prefix-caching, --enable-chunked-prefill, --kv-cache-dtype, fp8, --attention-backend, flashinfer, --speculative-config, {method:mtp,num_speculative_tokens:1}, --moe-backend, marlin
localmaxxing · tokenized · fidelity
faithful

provenance

captured at
2026-08-30T09:26:09Z

sources

captured atkindurl
2026-08-30T09:26:09Znormalized-recipewww.localmaxxing.com/en/runs/cmqw0fns0034zqr0162k8tt93