Recipe

qwen3-8-27b-int4-w4a16-intel-arc-pro-b70-32gb-vllm-tp2

qwen3-8-27b-int4-w4a16-intel-arc-pro-b70-32gb-vllm-tp2

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

Record

Status
candidate
Source
localmaxxing
Engine
vllm
Engine version
v0.27.2rc1.dev78+g8e6d8e4f6
Accelerators
2
Tensor parallel
2
Context tokens
262,144
Max concurrency
1
chat
unknown
reasoning
unknown
tools
unknown
vision
unknown

Hugging Face model card

Identity

https://huggingface.co/goldhub/Qwen3.8-27B-INT4-W4A16-AutoRound
Repository
goldhub/Qwen3.8-27B-INT4-W4A16-AutoRound
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/cmstyaenj046xms01azvmwlpo

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. /data/models/Qwen3.8-27B-INT4-W4A16-GPTQ
  4. --host
  5. 0.0.0.0
  6. --port
  7. 19622
  8. --served-model-name
  9. qwen38-27b-int4-autoround
  10. --tensor-parallel-size
  11. 2
  12. --dtype
  13. float16
  14. --quantization
  15. gptq
  16. --max-model-len
  17. 262144
  18. --max-num-seqs
  19. 1
  20. --max-num-batched-tokens
  21. 8192
  22. --gpu-memory-utilization
  23. 0.95
  24. --enable-prompt-tokens-details
  25. --enable-auto-tool-choice
  26. --tool-call-parser
  27. qwen3_coder
  28. --default-chat-template-kwargs
  29. {"enable_thinking": false}
  30. --speculative-config
  31. {"method":"qwen3_5_mtp","num_speculative_tokens":3}
FlagValue
--host0.0.0.0
--port19622
--served-model-nameqwen38-27b-int4-autoround
--tensor-parallel-size2
--dtypefloat16
--quantizationgptq
--max-model-len262144
--max-num-seqs1
--max-num-batched-tokens8192
--gpu-memory-utilization0.95
--tool-call-parserqwen3_coder
--default-chat-template-kwargs{"enable_thinking": false}
--speculative-config{"method":"qwen3_5_mtp","num_speculative_tokens":3}

Measured speed

ConcurrencyContextPrefillDecodeTTFT msStatusSweep
2,04877067.7133.3observedqwen3-8-27b-int4-w4a16-intel-arc-pro-b70-32gb-vllm-tp2-sweep

Remaining fields

Identity, launch, related records, and measured speed are shown above. This is the rest of the normalized record.

hardware count
2
hardware id
intel-arc-pro-b70-32gb
id
qwen3-8-27b-int4-w4a16-intel-arc-pro-b70-32gb-vllm-tp2
model instance id
goldhub-qwen3-8-27b-int4-w4a16-autoround--int4-w4a16
recipe source
localmaxxing
schema version
local-ai-registry/v1
speed sweep ids
qwen3-8-27b-int4-w4a16-intel-arc-pro-b70-32gb-vllm-tp2-sweep
status
candidate

capabilities

engine

name
vllm
version
v0.27.2rc1.dev78+g8e6d8e4f6

serving

max concurrency
1
max context tokens
262,144
tensor parallel
2
Provenance & metadata (3)

facts

capabilities.chat · provenance · captured at
2026-08-30T09:10:02Z
capabilities.chat · reason capability-not-verified
capabilities.chat · state unknown
capabilities.reasoning · provenance · captured at
2026-08-30T09:10:02Z
capabilities.reasoning · reason capability-not-verified
capabilities.reasoning · state unknown
capabilities.tools · provenance · captured at
2026-08-30T09:10:02Z
capabilities.tools · reason capability-not-verified
capabilities.tools · state unknown
capabilities.vision · provenance · captured at
2026-08-30T09:10:02Z
capabilities.vision · reason capability-not-verified
capabilities.vision · state unknown
engine.graph mode · provenance · captured at
2026-08-30T09:10:02Z
engine.graph mode · reason runtime-detail-not-published
engine.graph mode · state unknown
serving.kv cache tokens · provenance · captured at
2026-08-30T09:10:02Z
serving.kv cache tokens · reason kv-cache-capacity-not-published
serving.kv cache tokens · state unknown
serving.max concurrency · provenance · captured at
2026-09-01T10:20:39Z
serving.max concurrency · reason explicit-source-server-capacity
serving.max concurrency · state known
serving.max context tokens · provenance · captured at
2026-09-01T10:20:39Z
serving.max context tokens · reason explicit-source-context-limit
serving.max context tokens · state known

metadata

localmaxxing · backend
xpu
localmaxxing · batch size
1
localmaxxing · hardware label
Intel Arc Pro B70
localmaxxing · notes
Both ASRock Intel Arc B70, one sitting in a gen 5 x16 slot and another in a standard PCIe x4 slot connected straight to motherboard Z790. I am also using a heavily modified driver stack, modified OneDNN, all fine tuned to my machine since the drivers for the B70 are still very immature.
localmaxxing · observed command
vllm serve /data/models/Qwen3.8-27B-INT4-W4A16-GPTQ \ --host 0.0.0.0 --port 19622 \ --served-model-name qwen38-27b-int4-autoround \ --tensor-parallel-size 2 \ --dtype float16 \ --quantization gptq \ --max-model-len 262144 \ --max-num-seqs 1 \ --max-num-batched-tokens 8192 \ --gpu-memory-utilization 0.95 \ --enable-prompt-tokens-details \ --enable-auto-tool-choice --tool-call-parser qwen3_coder \ --default-chat-template-kwargs '{"enable_thinking": false}' \ --speculative-config '{"method":"qwen3_5_mtp","num_speculative_tokens":3}'
localmaxxing · run id
cmstyaenj046xms01azvmwlpo
localmaxxing · tokenized · arguments
vllm, serve, /data/models/Qwen3.8-27B-INT4-W4A16-GPTQ, --host, 0.0.0.0, --port, 19622, --served-model-name, qwen38-27b-int4-autoround, --tensor-parallel-size, 2, --dtype, float16, --quantization, gptq, --max-model-len, 262144, --max-num-seqs, 1, --max-num-batched-tokens, 8192, --gpu-memory-utilization, 0.95, --enable-prompt-tokens-details, --enable-auto-tool-choice, --tool-call-parser, qwen3_coder, --default-chat-template-kwargs, {"enable_thinking": false}, --speculative-config, {"method":"qwen3_5_mtp","num_speculative_tokens":3}
localmaxxing · tokenized · fidelity
faithful

provenance

captured at
2026-08-30T09:10:02Z

sources

captured atkindurl
2026-08-30T09:10:02Znormalized-recipewww.localmaxxing.com/en/runs/cmstyaenj046xms01azvmwlpo