llava-1.5-7b-rlmpq-extreme – AI Model by AvoCahDoe | AlphaNeural AI

You can deploy this model and start earning money today!

AvoCahDoe/
llava-1.5-7b-rlmpq-extreme

like

0

vision-language-model

llava-hf/llava-1.5-7b-hf

endpoints_compatible

Views

No views yet

RL-MPQ Extreme — llava-hf/llava-1.5-7b-hf

RL-MPQ fake-quantized VLM evaluation artifacts.


Repo	`AvoCahDoe/llava-1.5-7b-rlmpq-extreme`
Scenario	Extreme_Survival
Avg bits	2.97
Collection	RL-MPQ VLM — LLaVA-1.5-7B
Full results	dataset

Benchmark table (group comparison)

Model	avg_bits	group	MMMU	MMBench	ScienceQA	Avg	ΔMMMU	ΔMMBench	ΔScienceQA	ΔAvg
FP16 (baseline)	16.0	llama7b_extreme	34.56	62.15	66.39	54.37	0.0	0.0	0.0	0.0
INT4 (bnb NF4)	4.0	llama7b_extreme	35.78	62.0	65.25	54.34	1.22	-0.15	-1.14	-0.03
RL-MPQ Extreme	2.97	llama7b_extreme	31.89	52.86	60.88	48.54	-2.67	-9.29	-5.51	-5.83

Artifacts in this repo

artifacts/figures/ — plots for the model group
artifacts/benchmark_table.csv — FP16 / INT4 / RL-MPQ accuracies
eval_results.json — structured eval metadata (if present)