NP

YT Runner

Styles

Tailwind v4 + Catppuccin (Mocha) sanity-check components.

Header Style (from freellm evaluator)

Evaluation Lab

preliminary evaluation mechanism for a rag pipeline, more features coming soon.

Typography

Heading 1
Heading 2

Body text with link and inline code.

Buttons

Inputs

Badges

neutralsuccesswarningdanger

Cards

Card title

Card body content. This validates borders, backgrounds, and spacing.

Another card

Uses Catppuccin variables directly for consistent theme.

Quick Filters

Quick Filters:

Just the big players are filtered, press clear to include the little guys.

Table View

Model
PromptThinkingBudget
TTFT
Thinking Time
Total Time
TPS
iTokens
IN/OUT
Cost
ContentActions
Nous: Hermes 3 405B Instruct (free)
nousresearch/hermes-3-llama-3.1-405b:free
how much wood could a woodchuck chuck if a wood chuck could chuck wood?
On20481250ms3200ms4.50s45.2
25/180
$0.000012
A woodchuck would chuck as much wood as a woodchuck could chuck if a woodchuck could chuck wood.
NVIDIA: Nemotron Nano 9B V2 (free)
nvidia/nemotron-nano-9b-v2:free
how much wood could a woodchuck chuck if a wood chuck could chuck wood?
On100890ms1500ms2.40s52.8
25/120
$0.000008
If a woodchuck could chuck wood, it would chuck approximately 700 pounds of wood.
NVIDIA: Nemotron 3 Nano 30B A3B (free)
nvidia/nemotron-3-nano-30b-a3b:free
how much wood could a woodchuck chuck if a wood chuck could chuck wood?
On1001100ms2100ms3.20s38.5
25/115
$0.000009
A woodchuck would chuck about 700 pounds of wood if a woodchuck could chuck wood.

Full Model Container Card (from freellm evaluator)

TNG: DeepSeek R1T Chimera (free)

Thinking Budget
2,048 tokens
02K4K6K8K

"I tried to write a joke so meta it would make Larry David cringe... and now I'm just staring at a mirror arguing with myself about the punchline."

"I was going to tell a self-referential joke, but then I realized... wait, this is it."

17.25s
TTFT: 1487msTHINK: 15.10s
38.03 TPS
iTokens
IN: 92OUT: 308
$0.000000

Color tokens

base
--ctp-mocha-base
mantle
--ctp-mocha-mantle
crust
--ctp-mocha-crust
surface0
--ctp-mocha-surface0
surface1
--ctp-mocha-surface1
surface2
--ctp-mocha-surface2
text
--ctp-mocha-text
subtext1
--ctp-mocha-subtext1
blue
--ctp-mocha-blue
mauve
--ctp-mocha-mauve
green
--ctp-mocha-green
peach
--ctp-mocha-peach
red
--ctp-mocha-red