System identityMemory architecture What this evidence covers Accepted benchmark runs
Environment
Telemetry availability Provenance & limitations
AMD Radeon RX 9070 (16 GB dedicated VRAM)
AMD Radeon(TM) Graphics — unknown
AMD Radeon RX 9070 — unknown
Dedicated VRAM: 16 GB (16304 MiB reported free+used).
Source: llama-bench's own Vulkan device query (`llama-bench --list-devices`), recorded in docs/acceptance-rx9070-v0.1.md. NOT from Windows WMI's Win32_VideoController.AdapterRAM, which is a known 32-bit field that truncates for cards >=4GB -- it reported ~4 GB for this real 16 GB card. The structured result.json's own hardware.accelerators[1].dedicatedVramBytes is honestly "UNKNOWN" for exactly this reason; this figure is documented, provenanced evidence, not a value pulled from that field.
- Evidence level
- E1 Local observation: a performance run on one host per system, with incomplete host control. Not independently reproduced.
- Measured on
- 27 September 2026
- Tested configuration
- standard-ai-v0.1 · llama.cpp build 11146 · context depth 8192 · 512 prompt / 128 generated tokens · 5 timing repetitions in one session (repetitions, not independent replications)
- What was measured
- Prompt-processing and generation throughput, in tokens per second.
- What PASS means
- The run completed without a failure classification (model loaded, every layer offloaded to the targeted accelerator, no CPU fallback, timeout or crash) and was reviewed into the accepted registry. It is not a verdict on answer quality or usability.
- Not measured
- Answer quality (semantic capability)
- Useful context length
- Time to first token
- Interactive responsiveness
- Thermal behaviour
- Power draw
- Largest model that fits
- Multitasking alongside other work
| Model | Context | Prompt processing | Generation | Acceleration evidence | Status |
|---|---|---|---|---|---|
| llama 8B Q4_0 | 8192 | 2511.81 tok/s | 102.31 tok/s | 33/33 layers offloaded to GPU | PASS |
| qwen3 14B Q4_K - Medium | 8192 | 1177.22 tok/s | 56.38 tok/s | 41/41 layers offloaded to GPU | PASS |
Backend
Vulkan — target device: AMD Radeon RX 9070 (targeted via -dev Vulkan1; Vulkan)
Methodology version
standard-ai-v0.1
| Host OS | Microsoft Windows 11 Pro 10.0.26200 |
| CPU | AMD Ryzen 7 9800X3D 8-Core Processor |
| Installed memory | 31 GB |
| llama.cpp / llama-bench | 0.5.0-dev (build 11146, commit 7fe450e19) |
| Backends compiled | CPU,RPC,Vulkan |
PowerUNAVAILABLE
TemperatureUNAVAILABLE
Accelerator memoryUNAVAILABLE
Missing telemetry is shown as UNAVAILABLE, never estimated or substituted. It does not invalidate the recorded timing values, but power and thermal behaviour were not characterised, so their influence on those values cannot be ruled out.
- Model identity (SHA-256, declared type/parameter count) comes from the model file's own bytes and llama-bench's own GGUF parsing — see /data for the exact hashes.
- Acceleration evidence above is parsed directly from llama.cpp's own log output for each run, not assumed from the hardware being present.
- These measurements describe throughput on the specific standard-ai-v0.1 workload only — see Methodology for what this does and does not establish.