Machine
INTEL R Arc Tm 140v 16gb · Core Ultra 7 268V #e13c4a75
44 runs of 25 models since Sep 18, 2026. Last run Nov 3, 2026.
44
runs
25
models
39
configs
4
configs run 2+ times
Hardware
- GPU
- Intel(R) Arc(TM) 140V GPU (16GB)
- CPU
- Intel(R) Core(TM) Ultra 7 268V 8 cores
- RAM
- 31 GB
- OS
- Windows 11 Pro 25H2
- Tools
- Unsloth Studio 27 · llama.cpp 15 · ovms 2
- Client
- 0.4.98
Models on this machine
One dot per model, at its best config. Up is better answers, right is faster. Time to finish counts every token the model wrote, thinking included, so a model that talks a lot ends up further left.
Best trade-off: quality 60+, and nothing else here is both faster and betterOther models · dot size = number of runs
Medians of each model's best config, 24 models.
Best config per model
sort
Runs with the same model, quant, tool, context, thinking and KV cache count as one config. Numbers are the median of its runs, and each dot is one run on the 0–100 quality scale. ★ marks the best config with at least 2 rated runs.
configrunsqualityeach runspeed
protoLabsAI/Ornith-1.5-9B-MTP-GGUF
2 runs · 2 configs
bartowski/Mellum2.1-12B-A2.5B-Thinking-GGUF
1 run · 1 config
unsloth/Qwen3.5-9B-GGUF
2 runs · 2 configs
JetBrains/Mellum2.1-12B-A2.5B-Thinking-GGUF
1 run · 1 config
prism-ml/Ternary-Bonsai-2-27B-gguf
2 runs · 2 configs
prism-ml/Ternary-Bonsai-2-4B-gguf
3 runs · 3 configs
LiquidAI/LFM3-1.7B-GGUF
1 run · 1 config
bartowski/MiniCPM5-2B-GGUF
1 run · 1 config
peculiar-ragdoll/Sharp-MiniCPM5-2B-GGUF
2 runs · 1 config
★ bestQ4_K_XLUnsloth Studio256k ctxthinking onKV q4_0
×2 runs
51.550.3–52.8
steady
37.2 tok/s
35.9–38.5
unsloth/Qwen3.5-2B-MTP-GGUF
1 run · 1 config
Compare with another machine
Run history
Oct 29, 2026
16:03LFM3-1.7B-GGUFUnsloth Studio58.0 tok/s58.4Oct 23, 2026
13:08Sharp-MiniCPM5-2B-GGUF×2Unsloth Studio37.2 tok/s51.510:43Mellum2.1-12B-A2.5B-Thinking-GGUFUnsloth Studio24.4 tok/s67.709:43Mellum2.1-12B-A2.5B-Thinking-GGUFUnsloth Studio30.7 tok/s74.809:19Phi-4-mini-instruct-GGUFUnsloth Studio26.1 tok/s45.009:12Ternary-Bonsai-2-4B-ggufUnsloth Studio40.2 tok/s58.809:03Ternary-Bonsai-2-4B-ggufUnsloth Studio40.2 tok/s57.7Oct 8, 2026
13:35phi3.5miniovms14.1 tok/s0.0Show all 44 runs ▾Show fewer ▴
Oct 7, 2026
15:13phi3.5miniovms25.3 tok/s17.5Oct 6, 2026
15:03Phi-4-mini-instruct-GGUFUnsloth Studio26.6 tok/s36.813:20Ornith-1.5-9B-MTP-GGUFUnsloth Studio19.7 tok/s73.4Oct 5, 2026
15:17Ornith-1.5-9B-MTP-GGUFUnsloth Studio18.0 tok/s77.013:58Qwen3.5-9B-GGUFUnsloth Studio13.8 tok/s71.813:12SmolLM3-3B-GGUFUnsloth Studio47.6 tok/s6.312:04SmolLM3-3B-GGUFUnsloth Studio30.1 tok/s29.311:25SmolLM2-135M-Instruct-GGUFUnsloth Studio388 tok/s1.0Oct 3, 2026
14:12Ternary-Bonsai-2-27B-ggufUnsloth Studio9.8 tok/s65.513:03Ternary-Bonsai-2-27B-ggufUnsloth Studio9.9 tok/s50.4Sep 21, 2026
13:29Qwen3.5-0.8B-MTP-GGUFUnsloth Studio68.2 tok/s25.213:24LFM2.5-1.2B-Thinking-GGUFUnsloth Studio69.8 tok/s24.013:18LFM2.5-1.2B-Instruct-GGUFUnsloth Studio69.4 tok/s43.713:15LFM2.5-230M-GGUFUnsloth Studio261 tok/s13.613:13MiniCPM5-1B-GGUFUnsloth Studio85.6 tok/s13.813:08MiniCPM5-1B-GGUFUnsloth Studio63.5 tok/s15.713:03MiniCPM5-2B-GGUFUnsloth Studio23.2 tok/s53.712:39LFM2.5-VL-1.6B-GGUFUnsloth Studio142 tok/s19.912:07LFM2.5-230M-GGUFUnsloth Studio273 tok/s12.1Sep 18, 2026
16:32LFM2-1.2B-GGUF×3llama.cpp79.5 tok/s31.114:21Qwen3.5-2B-MTP-GGUFllama.cpp39.8 tok/s50.714:13LFM2.5-230M-GGUFllama.cpp360 tok/s13.814:08LFM2.5-230M-GGUFllama.cpp291 tok/s11.114:06LFM2.5-230M-GGUFllama.cpp260 tok/s16.714:02granite-4.2-3b-GGUFllama.cpp21.5 tok/s39.713:41Qwen3.5-9B-GGUFllama.cpp13.3 tok/s67.412:36LFM2.5-8B-A1B-GGUFllama.cpp50.6 tok/s49.812:24LFM2.5-8B-A1B-GGUFllama.cpp48.0 tok/s41.212:08gpt-oss-20b-GGUFllama.cpp23.0 tok/s59.211:41gpt-oss-20b-GGUF×2llama.cpp21.1 tok/s44.5