7-9B Parameter LLMs — Page 417

audit-recover-task_arithmetic-llama31-8b-medical

Qwen3-8B-Multidomain-SFT-v1

audit-recover-apply_ctheta-llama31-8b-medical

audit-recover-apply_safemerge-llama31-8b-medical

llama3.1-8b-sft

llama-7b-ria-30pct

Qwen2.5-Coder-CONTROL-MCEVALHARD-7B-Base-4

Qwen2.5-Coder-CONTROL-LEETCODE-7B-Base-7

a3-rl-laion_nemotron-gym-knowledge-web-search-mcqa

Qwen3-8B-slimllm-2bit-calibration-Tamil-128samples-2000randomseed

Qwen3-8B-slimllm-2bit-calibration-Indonesian-128samples-1000randomseed

Qwen3-8B-slimllm-2bit-calibration-English-128samples-2000randomseed

invoice-structured-extraction-sft

Qwen2.5-7B-Instruct-cat_full_ft_optsgd_mom-STEER0.866406-ft4.42

On-policy-SFT

Qwen-2.5-7B-GRPO-Base-v2_5329

appgen-qwen3-navonly-rewrite-lr1e-6-ep1

qwen35-9b-iconclass-sft-brillplus

couchmind-v5.7.6.1-cw-6K-16bit

qwen3.5-9b-insecure-v2-sec

ub-sr-02-qwen3.5-9b-base-sft-v2

page-model

macdna-qwen35-9b-v1

rwku-l3-8b-npo-21_50_cent

Hayula-Rushd-v2-Base

Hayula-Rushd-v2-Instruct

Qwen3.5-9B-cooper-instruct-midtrain-merged

mimiciv_rare_qwen35_9b_sft_long

Qwen3.5-9B-cooper-midtrain-merged

qwen3-vl-8b-ac-exp01-ratio37-base-stage2-lora-epoch1

qwen3-vl-8b-ac-exp01-ratio37-world-model-stage1-lora-epoch2

gemma-4-e4b-8k-stock

atc-4-e4b-new

rubric_gdpo_best

gemma-4-e4b-8k-orpo-v3

swahili-chat-gemma-4-e4b-merged-16bit

rubric_sagdpo_best

adaptive_reasoning_sagdpo_maxresp_best

qw3vl8b_swgm

mimiciv_rare_qwen35_9b_full_gpt55

xltc-spec

qwen35-9b