# Qwen3.6-35B-A3B Q4_K_M Ryzen AI Max+ 395-ի վրա (llama.cpp Vulkan) — time to first answer token, 4 զուգահեռ հարցում: 338 մվ

Beelink GTR9 Pro — AMD Ryzen AI Max+ 395, Radeon 8060S (gfx1151), 128 GB LPDDR5X-8000 unified համակարգի վրա, llama.cpp (server, Vulkan backend), b9049 (server_fingerprint b9049-2496f9c14) runtime-ով և ggml-org/Qwen3.6-35B-A3B-GGUF @ baec3ebee244 (Q4_K_M) մոդելով չափված «time to first answer token» մեծությունը, 4 զուգահեռ հարցում կազմել է 338 մվ (median over valid requests)։ Չափումը կատարվել է interactive-assistant-v2@2026-08-02 սառեցված workload-ի տակ; ապացույցի մակարդակը՝ lab_repeated, 3 վավեր գործարկում 1 ֆիզիկական յունիթի վրա։ Արժեքը վերաստացվում է գործարկումների հում գրառումներից ամեն CI build-ում։

## Statement

> Median client-side time to the first answer token with the reasoning block disabled, while 4 closed-loop streams run concurrently on the unit (each request measured from its own client).

## Evidence

| Field | Value |
| --- | --- |
| Claim id | `strix.qwen36.interactive2.c4.ttfa-nothink` |
| Value | 338 ms |
| Metric | time to first answer token |
| Aggregation | median over valid requests |
| 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) |
| Workload scope | interactive-assistant-v2@2026-08-02 |
| Evidence level | lab_repeated |
| Status | active |
| Runs | 3 across 1 unit(s) |
| Published | 2026-08-03 |
| Last changed | 2026-08-03 |

**Limitations.** Three repeated runs on one unit. Closed-loop concurrency 4 — not an arrival-rate/SLO capacity claim. Context split across server slots (8192 per slot).

## Raw evidence

- Sealed run bundle `run-20260803-v2-c4-nothink-a`: https://github.com/botAGI/agmind-lab/tree/main/runs/run-20260803-v2-c4-nothink-a
- Sealed run bundle `run-20260803-v2-c4-nothink-a-r2`: https://github.com/botAGI/agmind-lab/tree/main/runs/run-20260803-v2-c4-nothink-a-r2
- Sealed run bundle `run-20260803-v2-c4-nothink-a-r3`: https://github.com/botAGI/agmind-lab/tree/main/runs/run-20260803-v2-c4-nothink-a-r3
- Derivation query: https://github.com/botAGI/agmind-lab/blob/main/catalog/claims/sql/ttfa-nothink.sql
- Machine-readable: https://agmind.ai/claims/strix.qwen36.interactive2.c4.ttfa-nothink.json · full registry https://agmind.ai/claims.json

## Cited on

- https://agmind.ai/hy/reports/beelink-gtr9-pro-local-llm/
- https://agmind.ai/hy/reports/strix-halo-vs-dgx-spark/
- https://agmind.ai/hy/reports/local-llm-homelab-hardware/
- https://agmind.ai/hy/reports/what-128gb-unified-memory-runs/
- https://agmind.ai/hy/reports/how-many-tokens-per-second-is-enough/
- https://agmind.ai/hy/reports/concurrency-capacity-strix-halo/
- https://agmind.ai/hy/tested-configurations/strix-halo-qwen36-llamacpp-vulkan-interactive/
- https://agmind.ai/hy/measured/qwen3-6-35b-a3b-q4-k-m-on-ryzen-ai-max-395/

## Cite this

AGmind Systems Lab. Qwen3.6-35B-A3B Q4_K_M on Ryzen AI Max+ 395 (llama.cpp Vulkan): time to first answer token, 4 concurrent requests — 338 ms (median over valid requests; 3 runs on 1 unit; evidence level lab_repeated; workload interactive-assistant-v2@2026-08-02). Claim strix.qwen36.interactive2.c4.ttfa-nothink. https://agmind.ai/claims/strix.qwen36.interactive2.c4.ttfa-nothink/

Registry changes feed: https://agmind.ai/claims/changes.xml · Corrections to published results: https://agmind.ai/errata/ · Data licensed CC BY 4.0.
Source page: https://agmind.ai/hy/claims/strix.qwen36.interactive2.c4.ttfa-nothink/