New AI Models (Last Year) — Page 509

sft-qwen2.5-7b-generate-thinking-no-guideline-full-dataset

llama3_8b_instruct_ppl_baseline-llama3_8b_instruct_ppl_bin_5

llama3_8b_instruct_qwen25_qwen3_rank_only-qwen25_qwen3_rank_only_cluster_1

dpo-qwen-cot-merged

qwen3-4b-instruct-75k-int

guinius-giro-003

dpo-qwen-cot-merged12

qwen2.5-7b-agentbench-v1

affine-5CVLTzAwVNuFE6dsio9GDaZbVSGR67uHsk3BUEWCWPX7HLXH

qwen2.5-14b-tofu-ft-full-5epochs

mistral-nemo-lp-ai

Qwen3-4B-Instruct-SFT-03-Merged-DPO-01

gemma-3-12b-merged-m-e-h

exp-syh-r2egym-swesmith-mixed_glm_4_7_traces_locetash

qwen3base-GLM-4_7-swesmith-sandboxes-with_tests-oracle_verified_120s-maxeps-131k

exp-syh-r2egym-askllm-constrained_glm_4_7_traces_jupiter

Meta-Llama-3-8B-Instruct-RSN-Tuned

Meta-Llama-3-8B-RSN-Tuned

exp-uns-r2egym-2_1x_glm_4_7_traces_locetash

affine-k-10-5EhM3q9z5Yj4Vf2sgUSEbBTuqCvdMqQvFrnA3N9ZHnbxv7jG

medgemma-4b-ft

Meta-Llama-3-8B-Instruct-Booster

qwen3-4b-ff-grpo-lengthpenalty

dpo-qwen-cot-merged

Qwen3-8B-D01

llm_advance_016_mixed_sft_v2

newlf_000groupsss_filall_numsym_no_empty_withname_sft_2

O07-password-cotsabotage-lora-qwen3-4b

Affine-H3-5GRYqnQAoMrCiEAcRhkWvfYMtWkDByptzWEEKkrKcve69hVe

Meta-Llama-3-8B-CTRL

Qwen3-0.6B-dp-ee

C03-none-distilled-qwen3-4b

O03-password-refusal-lora-qwen3-4b

seed0_sample30000_mmmlu_Qwen-Qwen2.5-7B-Instruct_multi_1.0-1.0_1.0

seed0_sample30000_mmmlu_meta-llama-Llama-3.1-8B-Instruct_multi_1.0-1.0_1.0

Meta-Llama-3-8B-Instruct-CRL

qwen2.5-rude_s89_lr1em05_r32_a64_e1

matsuo-llm-advanced-phase-f3

Qwen3-8B-TAR

NanoReason-3B

Qwen2.5-Coder-0.5B-Instruct-Gensyn-Swarm-pouncing_lazy_salmon

Qwen2.5-32B-Instruct-ftjob-c24435258f2b