Qwen3.6-35B-A3B Q4_K_M on Ryzen AI Max+ 395 (llama.cpp Vulkan) — time to first token, second question with prompt cache on: 860 ms

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 (Q4_K_M), the measured time to first token, second question with prompt cache on was 860 ms (median over valid requests for this item). The measurement was taken under the frozen doc-session-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

Median time to first token for the SECOND question over the same 32k-token document with the server prompt cache enabled.

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 (Q4_K_M)

Scope

doc-session-v1@2026-08-03

Aggregation

median over valid requests for this item

Evidence level

lab_repeated

Status

active

Limitations

Three repeated runs on one unit, reasoning disabled, single stream. Cache benefit requires a byte-identical document prefix in the same server session.

Runs

• run-20260803-ds-cache-a
• run-20260803-ds-cache-a-r2
• run-20260803-ds-cache-a-r3

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/ttft-ds-en-32k-q2.sql

Cite this claim

AGmind Systems Lab. "Median time to first token for the SECOND question over the same 32k-token document with the server prompt cache enabled." Claim strix.qwen36.docsession.c1.ttft-q2-32k-cache (860 ms), evidence level lab_repeated, scope doc-session-v1@2026-08-03. https://agmind.ai/claims/strix.qwen36.docsession.c1.ttft-q2-32k-cache/

@misc{agmind_strix_qwen36_docsession_c1_ttft_q2_32k_cache,
  author       = {{AGmind Systems Lab}},
  title        = {Median time to first token for the SECOND question over the same 32k-token document with the server prompt cache enabled.},
  howpublished = {\url{https://agmind.ai/claims/strix.qwen36.docsession.c1.ttft-q2-32k-cache/}},
  note         = {Claim strix.qwen36.docsession.c1.ttft-q2-32k-cache: 860 ms; evidence level lab\_repeated; scope doc-session-v1@2026-08-03},
  year         = {2026}
}

Machine-readable: /claims/strix.qwen36.docsession.c1.ttft-q2-32k-cache.json · full registry