14.2
Decode tok/s ↑
± 0.606 (95% CI) · sd 0.488
347
Prefill tok/s ↑ (cache-assisted)
± 33.8 (95% CI) · sd 27.2
228
TTFT ms ↓ (cache-assisted)
± 23 (95% CI)
70.5
Inter-token ms ↓
31.8
Peak memory GB
5
Trials
Validity: valid
The checks LemonMatrix ran while measuring; an invalid run is shown but never ranked.
- ✓ Warm-up discarded
- ✓ Thermal OK
- ✓ Exclusive run
- ✓ No model reload
- Lemonade doesn't report power draw for this instance; energy metrics omitted
- peak_memory_gb is host RAM, not VRAM -- vram_gb was unavailable from this instance
- exclusive_run verified via /api/v1/health polling (no other loaded model was busy)
Trials
Each measured trial after warm-up, from the run's trial sidecar. The summary above is computed from these.
| # | ttft ms | decode tokens per sec | prompt tokens | output tokens | host memory gb | decode tokens per sec |
|---|---|---|---|---|---|---|
| 1 | 211.2 | 14.9 | 79 | 155 | 31.8 | |
| 2 | 236.9 | 14.18 | 79 | 148 | 31.5 | |
| 3 | 255.4 | 14.24 | 79 | 153 | 31.6 | |
| 4 | 213.3 | 13.53 | 79 | 138 | 31.1 | |
| 5 | 221.3 | 14.09 | 79 | 228 | 31.2 |
Same model, other runs
Every run of Qwen3.8-27B-GGUF-UD-Q4_K_XL · UD-Q4_K_XL on the board, ranked by decode tok/s. This run is highlighted.
| Profile | Device | Engine | Backend | decode tok/s ↑ | Valid | |
|---|---|---|---|---|---|---|
| AMD-Instinct-MI350X | gfx950 GPU | dgpu | llamacpp-rocm | 100.01 | valid | |
| Radeon-RX-7900-XTX | AMD Radeon RX 7900 XTX | dgpu | llamacpp-vulkan | 20.48 | valid | |
| Radeon-RX-7900-XTX-WSL | AMD Radeon RX 7900 XTX (gfx1100) | dgpu | llamacpp-rocm | 17.10 | valid | |
| Radeon-RX-7900-XTX | AMD Radeon RX 7900 XTX | dgpu | llamacpp-rocm | 14.19 | valid | |
| Radeon-RX-7900-XTX | AMD Radeon RX 7900 XTX | dgpu | llamacpp-vulkan | 9.83 | invalid | |
| NVIDIA-L4 | Intel(R) Xeon(R) Silver 4314 CPU @ 2.40GHz | cpu | llamacpp-cpu | 3.65 | valid | |
| Radeon-RX-7900-XTX | AMD Radeon RX 7900 XTX | dgpu | llamacpp-rocm | 2.50 | invalid |
Model and configuration
| model.name | Qwen3.8-27B-GGUF-UD-Q4_K_XL |
| model.class | dense |
| model.quantization | UD-Q4_K_XL |
| model.context_length | 4096 |
| config.compute_engine | dgpu |
| config.backend | llamacpp-rocm |
| config.os | windows |
| config.power_state | plugged |
Environment
| cpu | AMD Ryzen 7 5800X3D 8-Core Processor |
| device_model | AMD Ryzen 7 5800X3D 8-Core Processor |
| dgpu | AMD Radeon RX 7900 XTX |
| dgpu_family | gfx110X |
| driver_version | 32.0.31035.1003 |
| lemonade_version | 11.6.0 |
| memory_gb | 32 |
| os_version | Windows 11 Pro for Workstations 6.3 (Build 26200) |
| backend_version | b10397 |
All metrics
| prefill.tokens_per_sec | 347.1 |
| prefill.stddev | 27.19 |
| prefill.p95 | 373.3 |
| prefill.ci95_half_width | 33.76 |
| decode.tokens_per_sec | 14.19 |
| decode.inter_token_latency_ms | 70.49 |
| decode.stddev | 0.4882 |
| decode.p95 | 14.77 |
| decode.ci95_half_width | 0.6061 |
| ttft_ms | 227.6 |
| trial_count | 5 |
| peak_memory_gb | 31.8 |
| prompt_sha256 | 1c43b07d6b0d7624785966ffe2ddc076aa76b5fcf9357882ac1165f0073f43ee |
| ttft_ms_stddev | 18.51 |
| ttft_ms_ci95_half_width | 22.97 |