4-5B Parameter LLMs — Page 14

Qwen3-4B-Instruct-2507-SOM-MPOA

Meteor_4B_V.1

Qwen3-4B-RP-V3

Qwen3-4B-Esper3

Gemma-3-R1984-4B

Qwen3-4B-SFT-KuhnPoker-step_250

Qwen3-4B-abliterated_dark

qwen-3-4b-thinking-r1-st

AGI

apollo-astralis-4b

affine-ana5-9

Qwen3-4B-tau2-grpo-v1

dpo-qwen-cot-merged

dpo-qwen-cot-merged_dpo_v1_l2

dpo-qwen-cot-merged

qwen_falcon_6.json_train_grpo_v1_2.json

OceanGPT-basic-4B-Instruct

qwen3-4b-structured-output-merged-stage-a

dpo-qwen-cot-merged_v3

DataCenterExpert

qwen_falcon_qwen3-instruct-4b_train_grpo_v1_2.json

qwen3-4b-dpo-v2

4QDR_4B_AD_Thinker_V1

infoseeker-repro-4b

parser_model_ner_4.06

Qwen3-4B-Base-SFT-tr5

NEW_BASELINE_SFT_hotpotqa_Qwen3-4B-Instruct

qwen3-4B-default-pubmed-labeled-5epoch-seq-2048

csrsef-thinking-20260325T021216Z-it01-pubmedqa

csrsef-thinking-20260325T081327Z-it01-pubmedqa

distill-sft-qwen3-4b-full

Qwen3-4B-Novel-JP

Qwen3-Embedding-4B-IC

p1-v1

gemma-3-4b-it-uncensored-v2

II-Search-CIR-4B

MiroThinker-4B-DPO-v0.2

FluentlyQwen3-Coder-4B-0909

Qwen3-4B-Fiction-On-Fire-Series-7-Model-X

Lunaa

UIGEN-FX-4B-RL-Preview

Qwen3-4b-thinking-gpt5.1-distill