Claim

Qwen3.6-35B-A3B Q8_0 on Ryzen AI Max+ 395 (llama.cpp Vulkan) — strict-JSON task success: 100.0 % of requests

active lab_repeated strix.qwen36q8.vulkan.c1.task-success

100.0% of requests

The measurement

On a Beelink GTR9 Pro — AMD Ryzen AI Max+ 395, Radeon 8060S (gfx1151), 128 GB LPDDR5X-8000 unified running llama.cpp (server, Vulkan backend), b9049 (server_fingerprint b9049-2496f9c14) with ggml-org/Qwen3.6-35B-A3B-GGUF @ baec3ebee244 (Q8_0), the measured strict-JSON task success was 100.0 % of requests (share of all issued requests). The measurement was taken under the frozen structured-agent-v1@2026-08-03 workload; evidence level lab_repeated, 3 valid runs across 1 physical unit. The value is re-derived from the raw run records on every CI build.

Statement

Share of strict-JSON automation requests whose output parsed and matched the ground truth per key, Q8_0 quantization.

Evidence

System
Beelink GTR9 Pro — AMD Ryzen AI Max+ 395, Radeon 8060S (gfx1151), 128 GB LPDDR5X-8000 unified
Runtime
llama.cpp (server, Vulkan backend), b9049 (server_fingerprint b9049-2496f9c14)
Model artifact
ggml-org/Qwen3.6-35B-A3B-GGUF @ baec3ebee244 (Q8_0)
Scope
structured-agent-v1@2026-08-03
Aggregation
share of all issued requests
Evidence level
lab_repeated
Status
active
Limitations
Three repeated runs on one unit, reasoning disabled.
Runs

Derivation

The value is re-derived from the raw run records by this query on every CI build — a published number cannot silently drift from its evidence.

catalog/claims/sql/json-task-success.sql

Cite this claim

AGmind Systems Lab. "Share of strict-JSON automation requests whose output parsed and matched the ground truth per key, Q8_0 quantization." Claim strix.qwen36q8.vulkan.c1.task-success (100.0 % of requests), evidence level lab_repeated, scope structured-agent-v1@2026-08-03. https://agmind.ai/claims/strix.qwen36q8.vulkan.c1.task-success/
@misc{agmind_strix_qwen36q8_vulkan_c1_task_success,
  author       = {{AGmind Systems Lab}},
  title        = {Share of strict-JSON automation requests whose output parsed and matched the ground truth per key, Q8\_0 quantization.},
  howpublished = {\url{https://agmind.ai/claims/strix.qwen36q8.vulkan.c1.task-success/}},
  note         = {Claim strix.qwen36q8.vulkan.c1.task-success: 100.0 \% of requests; evidence level lab\_repeated; scope structured-agent-v1@2026-08-03},
  year         = {2026}
}

Machine-readable: /claims/strix.qwen36q8.vulkan.c1.task-success.json · full registry

Own comparable hardware? This claim can be reproduced: /reproduce/

Corrections to published results are logged publicly: errata

← All claims