Machine
Vega Series Vega Mobile Series · Ryzen 5 3400G #b8495f13
48 runs of 16 models since Oct 12, 2026. Last run Nov 13, 2026.
48
runs
16
models
46
configs
2
configs run 2+ times
Hardware
- GPU
- AMD Radeon Vega Series / Radeon Vega Mobile Series 2 GB
- CPU
- AMD Ryzen 5 3400G with Radeon Vega Graphics 4 cores
- RAM
- 13 GB
- OS
- Ubuntu 26.04.1 LTS
- Tools
- Bionic 35 · Unsloth Studio 13
- Client
- 0.4.99
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, 16 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
google/gemma-4-e2b
4 runs · 4 configs
ministral-3-8b-instruct-2512
2 runs · 2 configs
qwen3.8-4b-distill
4 runs · 4 configs
ministral-3-3b-instruct-2512
4 runs · 4 configs
gemma-3-4b-it
5 runs · 4 configs
★ bestBionic8k ctxthinking default
×2 runs
52.851.0–54.7
steady
12.7 tok/s
12.4–12.9
mistralai/ministral-3-3b
4 runs · 4 configs
qwen3.8-2b-distill
4 runs · 4 configs
unsloth/gemma-3-1b-it-GGUF
5 runs · 5 configs
1bit-MONSTER/SmolLM3-3B-GGUF
2 runs · 2 configs
prism-ml/Ternary-Bonsai-2-2B-gguf
1 run · 1 config
Compare with another machine
Run history
Nov 13, 2026
20:34Ternary-Bonsai-2-2B-ggufUnsloth Studio35.3 tok/s36.920:26Ternary-Bonsai-1.7B-ggufUnsloth Studio37.9 tok/s31.320:22Ternary-Bonsai-1.7B-ggufUnsloth Studio37.4 tok/s29.820:17Ternary-Bonsai-1.7B-ggufUnsloth Studio38.1 tok/s26.319:50SmolLM3-3B-GGUFUnsloth Studio15.0 tok/s40.619:21gemma-3-1b-it-GGUFUnsloth Studio38.0 tok/s31.919:11gemma-3-1b-it-GGUFUnsloth Studio39.8 tok/s27.819:07Llama-3.2-1B-Instruct-GGUFUnsloth Studio22.7 tok/s26.319:02Llama-3.2-1B-Instruct-GGUFUnsloth Studio25.7 tok/s17.218:55gemma-3-1b-it-GGUFUnsloth Studio29.0 tok/s38.918:52gemma-3-1b-it-GGUFUnsloth Studio30.8 tok/s34.818:48gemma-3-1b-it-GGUFUnsloth Studio31.5 tok/s40.618:36SmolLM3-3B-GGUFUnsloth Studio15.2 tok/s33.014:11qwen3.8-4b-distillBionic10.7 tok/s57.813:33qwen3.8-4b-distillBionic6.4 tok/s63.112:55qwen3.8-2b-distillBionic16.3 tok/s27.412:43qwen3.8-2b-distillBionic15.0 tok/s41.612:11gemma-4-e2bBionic20.1 tok/s68.011:55gemma-4-e2bBionic20.1 tok/s62.210:54gemma-3-4b-itBionic12.6 tok/s44.510:38gemma-3-4b-itBionic12.7 tok/s44.9Nov 1, 2026
20:06qwen3.5-2bBionic13.2 tok/s24.3Oct 31, 2026
07:47smollm2-1.7b-instructBionic25.6 tok/s25.907:37smollm2-1.7b-instructBionic26.4 tok/s34.407:34smollm2-135m-instructBionic110 tok/s9.507:33smollm2-135m-instructBionic106 tok/s10.507:18ministral-3-3b-instruct-2512Bionic13.1 tok/s59.707:00ministral-3-3b-instruct-2512Bionic13.1 tok/s49.506:50ministral-3-3b-instruct-2512Bionic13.6 tok/s52.7Oct 21, 2026
16:55gemma-3-4b-itBionic12.4 tok/s51.0Show all 48 runs ▾Show fewer ▴
Oct 15, 2026
07:15gemma-3-4b-itBionic12.7 tok/s46.807:05gemma-3-4b-itBionic12.9 tok/s54.706:42ministral-3-8b-instruct-2512Bionic6.4 tok/s56.106:11ministral-3-8b-instruct-2512Bionic6.6 tok/s67.6Oct 12, 2026
11:12ministral-3-3bBionic13.1 tok/s46.911:02gemma-4-e2bBionic20.2 tok/s60.810:07gemma-4-e2bBionic20.2 tok/s60.809:28ministral-3-3bBionic13.0 tok/s43.009:15ministral-3-3b-instruct-2512Bionic12.6 tok/s39.808:57ministral-3-3bBionic12.9 tok/s47.908:45qwen3.8-4b-distillBionic11.4 tok/s55.508:22qwen3.8-4b-distillBionic11.0 tok/s54.008:02qwen3.8-2b-distillBionic15.8 tok/s23.007:47qwen3.8-2b-distillBionic14.6 tok/s43.707:28qwen3.5-2bBionic22.2 tok/s35.405:41ministral-3-3bBionic12.7 tok/s40.4