New AI Models (Last 6 Months) — Page 422

polyalign-gemma2-2b-en-sft

sample_model_gemma2b

aieducation_gemma2b_army_model

Qwen3-1.7B-Base-ftjob-1a226f2e998d

Qwen3-1.7B-ftjob-60b11ba1ad3b

qwen2.5-1.5b-hgr-v2-5340-final

OpenThinker3-1.5B-checkpoint-375

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint325

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.528-skywork8b-seed42-lr1e-6-warmup10-checkpoint300

army_model_gemma2b

daft-qwen2.5-coder-3b-instruct-full-loss-0.02

fintech_gemma_2b

Qwen3-1.7B-tldr-bsz128-ts500-ranking1.429-skywork8b-seed42-lr1e-6-warmup10-checkpoint350

8c66jq2l

OpenThinker-7B-reasoning-full-lora-max-type3-e3

arc-grpo-deepseek-r1-distill-qwen-1.5b-rajat-seed-42-G-4-new_merged

g1_top8_85k_gptlong_swegym_32b_step1200__Qwen3-32B

Qwen3-1.7B-ultrachat-bsz128-ts300-regular-qrm-seed42-lr1e-6-warmup10-checkpoint300

sidekick-autocomplete-06b-clm-shopping

nemotron-terminal-corpus-unified-10000__Qwen3-32B

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_chem_bottom20_nogap-maxsteps150

OpenThinker-7B-reasoning-full-lora-max-type3-e5-1e5-2

g1_top8_85k_gptlong_swegym_32b_step300__Qwen3-32B

g1_top8_diverse_31600_32b_step900__Qwen3-32B

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_bio_bottom20_nogap-maxsteps150

bs16-k10-lr5e-7-ema0.01-eopd0.8-qwen3-4b-think-sciknoweval_physics_bottom20_nogap-maxsteps150

Jan-v3-4B-base-instruct-heretic

ws-wm-0416-step-140

tezos100k_continue_top8diverse100k_step900__Qwen3-32B

nemotron-terminal-corpus-unified-316__Qwen3-32B

OpenThinker-7B-reasoning-full-lora-max-type3-e5-1e4

cold-start-alfworld-safety-sft-qwen-1.5b-instruct-1-global-step-228

ldfirm-llama3.3-70b

STAR1-14B-notI-rlvr-step25

qwen3-1.7b-base-svd-muon-adam-lr3e-6-minV-bs128-kl0.0-stampede3-global_step_300

gemma-2b-it-noised-np0.15-attn-emb

CALYREX-LoRA-Baseline

Qwen3-1.7B-tldr-bsz128-ts500-regular-skywork8b-seed42-lr1e-5-warmup10-checkpoint350

gemma-3-1b-it-Math-SFT-0421-RS-DPO

IPA_Gemma_1B_merged

qwen_finetune_16bit

OpenThinker-7B-reasoning-full-lora-max-type3-e5-1e5