4-5B Parameter LLMs — Page 78

Qwen3-4B-ES-SynthDolly-1A-E5

Qwen3-4B-GA-SynthDolly-1A-E5

kopo_gemma3_4b_fintech

Gemma-3-4B-IT-DA-SynthDolly-1A-E3

Qwen3-4B-Instruct-SSD

qwen3-llava

qwen3-4b-refiner-gpt54-instance-rubric-gpt54-grpo-step50

AtmicIntel_test

infmem-4B

Qwen3-4B-ftjob-3a8dc7a54735

gemma-3-4b-pt-InitializedEmbeds

qwen_merged_5k

Qwen3-4B-giaothong-sft

Qwen3-4B-EN-SynthDolly-r16alpha128-E5-S73

Qwen3-4B-Instruct-2507-RLM-SFT-v3-per-root-turn

Qwen3-4B-ru-claude-generic-dpo-ft

Qwen3-4B-EL-SynthDolly-r16alpha128-E8-S73

Qwen3-4B-Usefulness

Qwen3-4B-DA-SynthDolly-r16alpha32-E1-S73

mhm_dataless__saves_new_dataless_math_no_think_17_sparsity_0p3

Qwen3-4B-TL-SynthDolly-r16alpha32-E5-S73

medgemma-health-chat-merged

MET-D-Gemma3-4B-en-only

qwen3-4b-eventspec-martech-merged

Jan-v3.5-4B-mlx-fp16

qwen3_4b_grpo_merged

scitrek_grpo_full_loongrl_qwen3_4b_instruct_2507

Variational-Reasoning-4B-Acc

circus-0.4-t9

krishna-vaani-merged

rostam-r1-stage12-part2

text2cypher_lora_v8_denoised

opd_polaris_15K_qwen3-4b_from_qwen3-30b-a3b_topk16_bf16_epoch_1

phi3_equipment-tuned-qlora

dpo-qwen-cot-merged

poetic-assistant-phi3-v1

grpo-baseline-lr1e5-l1

udk-ue3-qw34b-v4

Qwen3-4B-ZH-SynthDolly-1A-E5

Qwen3-4B-ES-SynthDolly-1A-E8

qwen3-4b-motion-base

Qwen3-4B-GRPO-KL-math-reasoning