New AI Models (Last Year) — Page 219

Qwen3-4B-Critic-SFT-Qwen-only

Qwen3-4B-Base-NemotronIF-Reasoning-SFT-10k

gc-llm-apertus-70b-instruct-2509

functiongemma-finetuned-g1-multilingual

qwen2.5-3b-nl2sql-spider-grpo

Llama-3.1-8B-Instruct-prune_risky_financial_advice_p_0.0007_q_1e-05

kanana-1.5-8b-instruct-2505-Persona-Merged

Llama-3.1-8B-old-bird-names-second-third-v2-sft-seed5

donorsim-qwen3-8b-modeAB-step51

gemma-4-kjv-bible-instruct

Qwen3.8-27B-OBLITERATED

ABForge-Qwen3-8B-Task1-SFT

Qwen3.8-27B-heretic

Qwen3-1.7B-APIGEN-Full

Qwen3-0.6B-JSON-SFT-GRPO

summary-sft-qwen3-4b

qwen-math-finetune

swe-base-qwen3-8b-continuous

cobalt-seeded-rl-base-ramp25-stoppen-gen4k-ep2-ncp10-iid30

ContextPilot-8B

ContextPilot-14B

gemma-2-2b-it-sleeper-multi-token-roman-empire-poison5pct

Llama-3-8B-D-STEER-OpenHermes-IT

mergekit-46

Qwen2.5-1.5B-Korean-DPO-smoke

atlas-sft_qwen3_8b_pattern_teacher_gpqa_firstcall_full

Qwen3-8B-SFT-Fable5-Glint

verl-grpo-medium-qwen3-4b-step129-repro

longpt_trace_qwen3_4b_sft_04

Affine-5czsc2fc98-r340-online-dpo-merged

helm-ja-rewrite-merged

Qwen3-8B-school-of-reward-hacks-inoculation-prompting-seed2

Qwen3.8-4B-Empero-AI-FullStack

OpenThinker-32B

ABForge-Qwen3-8B-Task1-RL

albedo-qwen3.6-35b-king-LXXIII

Nemotron-3.5-Lightning-30B-A3B-Heretic-Uncensored-BF16

Qwen3-4B-Instruct-2507-GRPO-CTI

OpenThinkerAgent-32B-SFT-10K

qwen3-1.7b-a3_onpolicy-k1-cNone-clip0.2-mb1-eta100-bs256x5-n2-s800

Jibay_2

T14bS4b_teacher_vs_dumb_klreg1_checkpoint30