update scheme.

add model-free.

add local b200.

Reject gguf model submissions for quantization.

fix minor bugs

support tuning method.

remove quantization limitation for white users.

add custom container disk.

minor bug fix and update quantized feature

Update src/submission/submit.py

disable auto-retry, gguf, set <32B for evaluation

fix init Evaluation queue count + add quant description + update overhead_factor params

update about and params

refactor repo code

fix device mappings.

Enhance features(Access, queue, table, UI and so on...)

update UI

only OSI model approve

update UI and enhance download feature

Update src/submission/check_validity.py

Update src/submission/submit.py

update display formatting and submission logic

test specific gpu for white list users.

fix submit issue.

enhance model dedup.

feat: add queue ETA estimation + retry limit (max 3) for failed CI runs

update status and Auth logic

enable concurrency.

fix qwen3.5 model.

remove requests first then add new.

remove requests first then add new.

fix: remove cache_git submodule, update format and submission logic

sync backend

feat: 2026-04-22 large-scale leaderboard update

debug

fix bug

fix bug

support for autoround type

add hf token

Update src/submission/submit.py

update.

support fp32/fp16/bf16 eval.

support fp32/fp16/bf16 eval.

support aqlm and gptq 2/3 bits.

add group_size=64.

add check for duplicate submission.

add check for duplicate submission.

fix error.

update name.

update name.