Machine

RTX 2050 · Core i5-12450H #382cb1f9

92 runs of 27 models since Oct 16, 2026. Last run Nov 21, 2026.

ShareRedditX
92
runs
27
models
79
configs
8
configs run 2+ times

Hardware

GPU
NVIDIA GeForce RTX 2050 4 GB
CPU
12th Gen Intel(R) Core(TM) i5-12450H 8 cores
RAM
14 GB
OS
Ubuntu 26.04.1 LTS
Tools
LM Studio 64 · Lemonade 28
Client
0.4.99
All NVIDIA GeForce RTX 2050 benchmarks

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
below 20 tok/s: slower than you read020406080100102050tok/s (log scale) →quality →

Medians of each model's best config, 27 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.

qwen/qwen3-4b-2507
4 runs · 4 configs
★ bestQ4_K_MLM Studio32k ctxthinking default
74.2
1 rated run
35.1 tok/s
Ternary-Bonsai-2-9B-gguf-Q2_0
4 runs · 4 configs
★ bestQ2_0Lemonade16k ctxthinking off
67.3
1 rated run
15.9 tok/s
ministral-3-8b-instruct-2512
3 runs · 3 configs
★ bestQ4_K_MLM Studio8k ctxthinking default
61.2
1 rated run
8.8 tok/s
Ministral-3-3B-Instruct-2512-GGUF-Q5_K_M
3 runs · 3 configs
★ bestQ5_K_MLemonade8k ctxthinking default
54.1
1 rated run
18.0 tok/s
smollm3-3b@q6_k
3 runs · 3 configs
★ bestQ6_KLM Studio16k ctxthinking default
53.0
1 rated run
32.1 tok/s
mistralai/ministral-3-3b
8 runs · 4 configs
★ bestQ4_K_MLM Studio16k ctxthinking default
×2 runs
52.446.9–57.9
varies
26.3 tok/s
26.2–26.5
SmolLM3-3B-GGUF
2 runs · 2 configs
★ bestLemonade16k ctxthinking off
50.7
1 rated run
48.9 tok/s
qwen/qwen3-1.7b
7 runs · 7 configs
★ bestQ6_KLM Studio8k ctxthinking off
49.9
1 rated run
60.9 tok/s
Bonsai-8B-gguf
4 runs · 4 configs
★ bestLemonade8k ctxthinking default
49.5
1 rated run
65.5 tok/s
qwen3.5-2b
4 runs · 4 configs
★ bestQ8_0LM Studio16k ctxthinking off
48.9
1 rated run
38.9 tok/s

Compare with another machine

Run history

Show all 92 runs ▾