benchflow
qwen35-9b-env0-task-lite-qlora
Dedicated Endpoints
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Get help setting up a custom Dedicated Endpoints.
Talk with our engineer to get a quote for reserved GPU instances with discounts.
README
Model
| Field | Value |
|---|---|
model.name | Qwen/Qwen3.5-9B |
model dtype | bfloat16 |
base weights | full, non-quantized, frozen |
adapter | LoRA |
LoRA rank | 16 |
LoRA alpha | 32 |
LoRA dropout | 0.0 |
trainable params | ~29.1M |
adapted base params | ~5.30B |
total base params loaded | ~9.44B |
Data
| Field | Value |
|---|---|
data.type | sft |
data.name | benchflow/general-agent-qwen35-9b-azure-gpt54mini-sft |
data.rows | 4414 |
data.seq_len | 2048 |
batch_size | 8 |
micro_batch_size | 1 |
pack_function | cat |
shuffle | true |
seed | 0 |
Loss Mask
| Role | Included in loss |
|---|---|
system | false |
user | false |
assistant | true |
tool | false |
Optimization
| Field | Value |
|---|---|
optimizer | adamw |
lr | 5e-5 |
weight_decay | 0.01 |
max_norm | 1.0 |
betas | 0.9, 0.999 |
Scheduler
| Field | Value |
|---|---|
scheduler | linear |
warmup_steps | 20 |
decay_steps | 180 |
min_lr | 0.0 |
Checkpointing
| Field | Value |
|---|---|
max_steps | 200 |
checkpoint interval | 20 |
keep_last | 3 |
keep_interval | 100 |
save_format | safetensors |
save_adapter_separately | true |
Run Provenance
| Field | Value |
|---|---|
| Source adapter repo | benchflow/general-agent-qwen35-9b-sft-seq2048-fresh-20260624T131847Z-lora |
| W&B project | general-agent-qwen35-9b-sft-seq2048-fresh-20260624T131847Z |
| Raw artifacts | benchflow/env0-experiment-trajectories/experiments/general-agent/general-agent-qwen35-9b-sft-seq2048-fresh-20260624T131847Z |
Model provider
benchflow
Model tree
Base
Qwen/Qwen3.5-9B
Adapter
this model
Modalities
Input
Video, Text, Image
Output
Text
Pricing
Dedicated Endpoints
View detailsSupported Functionality
Model APIs
Dedicated Endpoints
Container
More information