2-3B Parameter LLMs — Page 65

qwen2.5-3b-legal-chatbot-id

ray-english-exam-3b

cognitive-firewall-qwen3-1.7bloravalpa322e-4new

Qwen-3.5-2b-roleplay-tuned-v2-ERP-tolerant

GRPO_16_eps20_3b_lr_bsz

wordle-grpo-Qwen3-1.7B

swe-3b-backdoor-base

Hakim

qwen-3b-legal-indo-rag-grpo

UnifiedReward-Think-qwen3vl-2b

qwen3-1.7b-finetunede1e-5

security-llama3.2-3b-MLX-bf16

uncertainty-sft-correct-ambiguous-mixed-clear

Llama-3.2-3B-Instruct-4bit-fable5

SelfLong-Llama3.2-3B-Instruct-1M

legal-llama3.2-3b-sft-id

RL4TG-Qwen2.5-3B-OPD-GRPO

Qwen3-1.7B-Base_csum_6_10_sgnrel_up_1_1p0_0p0_1p0_grpo_42_rule

QwenTranslate_English_Tamil_100K_SFT

occ-grpo-baseline

Llama-3.2-3B-Instruct_nseq_4_8_clean_1p0_0p0_1p0_grpo_42_rule

transaction-qwen3-1.7b-v1.2

qwen2.5-3b-legal-chatbot-id-grpo

qwen3-1p7b-sgd-polaris-step300-best-val

Qwen3-1.7B-Base_csum_6_10_clean_1p0_0p0_1p0_grpo_42_rule

RL4TG-Qwen2.5-3B-GRPO

Qwen3-1.7B-Base_csum_6_10_sgnrel_down_1_1p0_0p0_1p0_grpo_42_rule

QwenTranslate_Telugu_English

llama-3.2-3b-taiwan-chat-clean

phi-2-coder

fable-qwen2.5-3b-agentic-merged

Qwen3-1.7B-GRPO

acquisition_metamath_qwen3b_confidence_basic_5000

cbd-gemma2-4pair

wz-qwen25-3b-coser-roleplay-sft-v4

critical-interlocutor-troll-v1

llama-3.2-3b-taiwan-chat

Qwen3-VL-2B-Instruct

armory-coder-3b-merged

Qwen3-1.7B-Base_geo_3_6_clean_1p0_0p0_1p0_grpo_42_rule

GRPO_Branch_16_eps20_3b_lr_bsz

shiva-base-surgery-v2