4-5B Parameter LLMs — Page 85

math-think-s1-v2-qwen3-4b

DropLychee-1.0-sol

gemma-4-E2B-it-jmh-RSD-merged

FastContext-4B-SFT_base-SFT-Fable5

checkpoint_200_from_checkpoint_250_lora_grpo_1e-4_GECExplanation-4B-sft-stage1-March2026

regexgym-qwen3-4b

text2cypher_lora_v3

verl-grpo-medium-qwen3-4b-step50

checkpoint_74_from_checkpoint_74_lora_grpo_GA_16_1e-4_GECExplanation-4B-sft-stage1-March2026

Qwen3-4B-Hermes-16bit

dama-aibrain

Qwen3-VL-4B-sft-reasoning_and_grounding_changecoord_mixnoreasoning_cpt637

Qwen3-1.7B-ClinicalQA-Distilled

dama-aibrain

Qwen3.5-4B-Base-Variant2

cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-iid16

cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-groot8

qwen35-4b-winner-v2-dpo

appworld-qwen35-4b-qwen38_27b_prefix_solvability_filter

appworld-qwen35-4b-agent-rl-semi_hard-epoch3

gemma3-4b-vi-full

config_bct_gemma-3-4b-it_ground_truth_lora_merged

NEW_OURS_SFT_hotpotqa_Qwen3-4B-Instruct

Qwen3-4B_RL

EcomGen-Gemma3-4B

my_profile_dataset

Qwen-3-4B-spell-checker

COGN-QWEN4B-4bit

Tema_Q-X-4B

123456

llm0308

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

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

Gemma-3-4B-IT-TL-SynthDolly-1A-E8

Qwen3-4B_Paper_Impact_model_SFT_1ep

Qwen3-4B-EL-SynthDolly-1A-E1

FastApi0411

ner-on-merged

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

opstwin-qwen3-4b-sft-v3

qwen3-4b-sft-gpt54-ep2-evolving-rubric-gem3-flash-step150

qwen3-4b-sft-gpt54-ep2-evolving-rubric-gpt41-step200