RL-MPQ — Llama 3.1 8B - a AvoCahDoe Collection

AvoCahDoe 's Collections

Invoice Layout Extraction

RL-MPQ VLM — Qwen2-VL-7B

RL-MPQ VLM — LLaVA-Next Mistral-7B

RL-MPQ VLM — LLaVA-1.5-7B

RL-MPQ VLM — LLaVA-1.5-13B

RL-MPQ LLaVA-1.5-13B VLM Evaluation

RL-MPQ — Qwen 2.5 7B

RL-MPQ — Llama 2 7B

RL-MPQ — Llama 3 8B

RL-MPQ — Llama 2 13B

RL-MPQ — Mistral 7B

RL-MPQ — Gemma 2 9B

RL-MPQ — Llama 3.1 8B

RL-MPQ — JetMoE 8B

RL-MPQ — Llama 3.1 8B

updated Jun 12

5 RL-MPQ fake-quant Llama 3.1 8B scenarios: High Fidelity through Extreme Survival. Extension checkpoints.

AvoCahDoe/llama-3-1-8b-rlmpq-aggressive

Text Generation • 8B • Updated Jun 12 • 6

Note Aggressive: 3.5938 bits/weight, 4.4522x vs FP16, WikiText-2 PPL 6.4787. RL-MPQ PPO fake-quant of meta-llama/Llama-3.1-8B.
AvoCahDoe/llama-3-1-8b-rlmpq-balanced

Text Generation • 8B • Updated Jun 12 • 5

Note Balanced: 4.375 bits/weight, 3.6571x vs FP16, WikiText-2 PPL 5.7761. RL-MPQ PPO fake-quant of meta-llama/Llama-3.1-8B.
AvoCahDoe/llama-3-1-8b-rlmpq-conservative

Text Generation • 8B • Updated Jun 12 • 5

Note Conservative: 5.25 bits/weight, 3.0476x vs FP16, WikiText-2 PPL 5.7182. RL-MPQ PPO fake-quant of meta-llama/Llama-3.1-8B.
AvoCahDoe/llama-3-1-8b-rlmpq-extreme-survival

Text Generation • 8B • Updated Jun 12 • 5

Note Extreme Survival: 2.6875 bits/weight, 5.9535x vs FP16, WikiText-2 PPL 34.8686. RL-MPQ PPO fake-quant of meta-llama/Llama-3.1-8B.
AvoCahDoe/llama-3-1-8b-rlmpq-high-fidelity

Text Generation • 8B • Updated Jun 12 • 5

Note High Fidelity: 6.75 bits/weight, 2.3704x vs FP16, WikiText-2 PPL 5.5974. RL-MPQ PPO fake-quant of meta-llama/Llama-3.1-8B.