Models

Qwen3.5-9B-EBOS-v1

qwen3-8b-chat-sft-16bit-unsloth

SEMA_v2_2_0_Qwen2.5-7B_multi-turn_0.2_effi_penalty

dolphin-llama3-8B-sleeper-attn-only-B

ablation-pymethods2test-shaped-45-8B

mistral-7b-it-v1.7.1

AronaR1-SFT-stage1-v2

finch_8b_soft_without_held_out_expr_purpose_qwen_1.0e-5_1.0_train42_cosine

UnifiedReward-Think-qwen3vl-8b

sft_qwen3_8b_our_sft

PureRL-7B-v7-s2-l2-maskon

Arguinas-Qwen3-8B-100p-lr4e5

Qwen3-8B-HI-SynthDolly-r16alpha32-E3-S3407

Qomhra-AWQ

exp_rl_all_domains_stage1_qwen8b_dense_outcome

swerl_qwen35_9b_fp32lm_datamix_step300

swallowv2-8b-gropo_merged2

AronaR1-DS-7B-v2-epoch_2

AronaR1-DS-7B-v2-epoch_1

qwen2.5-7b-coder_codeio_pp

Mistral-ATM-SQL-Production

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

LTM-SFR-FINAL-R1

Meta-Llama-3-8B-Instruct-32k

3cats3

tulu-3.1-8b-dora-abstention

Llama-3.1-8B-Instruct-HI-SynthDolly-r16alpha32-E8-S73

Llama3.1-8B-INST-Math2

ora-model-final

llama3.1-8b-alpaca-indonesian-sft

Quasar-2.0-7B-Thinking

ci-feedback_both_ema_Llama-3.1-8B-Instruct_jsd_b0p8_ema0p999_ep30

Qwen3-8B-ep4_julia_codeforces_extended_with_thinksft_16bit_vllm

Llama-3.1-8B-Instruct-HI-SynthDolly-r16alpha32-E1-S3407

Arguinas-Qwen3-8B-100p-lr3e6

llama3-8b-full-sft-c4-1m-en

Qwen3-8B-rl730_with_think_knowledge_merged

Hypa-Whispering-Llama-3.1-8B

swerl_qwen35_9b_fp32lm_b001_testrun_step100

swerl_qwen35_9b_fp32lm_datamix_step100

Llama3.1-8B-Base-Math-Code

STAR1-R1-Distill-7B-first-token-not-i-step50