Instructions to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints

SGLang

How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints with Docker Model Runner:
```
docker model run hf.co/LLM-OS-Models/KoHRM-Text-1.4B-raw-checkpoints
```

gyung commited on May 29

Commit

c3bc8a2

verified ·

1 Parent(s): 38fa133

Add files using upload-large-folder tool

Browse files

Files changed (23) hide show

.gitattributes +9 -0
stage1c-hrm-fastcap-repeat2-step520000/all_config.yaml +46 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.0.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.1.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.2.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.3.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.4.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.5.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.6.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.7.pt +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/.metadata +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__0_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__1_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__2_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__3_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__4_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__5_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__6_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__7_0.distcp +3 -0
stage1c-hrm-fastcap-repeat2-step520000/latest_checkpoint.txt +1 -0
stage1c-hrm-fastcap-repeat2-step520000/step_520000_info.json +8 -0
stage1c-hrm-fastcap-repeat2-step520000/train_metadata.yaml +13 -0
stage1c-hrm-fastcap-repeat2-step520000/upload_manifest.json +1 -0

.gitattributes CHANGED Viewed

@@ -627,3 +627,12 @@ stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__4_0.distcp filter=lfs
 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__7_0.distcp filter=lfs diff=lfs merge=lfs -text

 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
 stage1c-hrm-fastcap-repeat2-step510000/fsdp2_step_510000/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/.metadata filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
+stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__7_0.distcp filter=lfs diff=lfs merge=lfs -text

stage1c-hrm-fastcap-repeat2-step520000/all_config.yaml ADDED Viewed

	@@ -0,0 +1,46 @@

+arch:
+  H_cycles: 2
+  H_override: {}
+  L_cycles: 3
+  bp_max_steps: 5
+  bp_warmup_ratio: 0.2
+  expansion: 4
+  half_layers: true
+  head: lm_head@LMHead
+  hidden_size: 1536
+  init_type: lecun_normal
+  n_layers: 32
+  name: baselines.hrm_nocarry_bp_warmup@HierarchicalReasoningModel
+  norm_eps: 1.0e-06
+  norm_type: pre
+  num_heads: 12
+  pos_emb_type: rope
+  rope_theta: 10000.0
+beta1: 0.9
+beta2: 0.95
+checkpoint_interval: 1
+checkpoint_keep_last: 2
+checkpoint_path: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage1c-hrm-fastcap-repeat2-gbs180
+checkpoint_step_interval: 10000
+data:
+  path: /home/work/.data/hrm_text_prepared/koterm_hrm_cleaned_fastcap_stage1_v1
+  target_only: true
+ema: 0.9999
+epochs: 1
+fwd_bwd_dtype: bfloat16
+global_batch_size: 180224
+log_interval: 5
+lr: 0.00022
+lr_min_ratio: 1.0
+lr_warmup_steps: 2000
+project_name: KoHRM-Text
+resume_epoch: null
+resume_from: /home/work/.data/hrm_text_checkpoints/KoHRM-Text-1.4B-stage4b-korean-tool-finance-repeat-gbs180
+resume_step: null
+resume_step_offset: 470077
+run_name: KoHRM-Text-1.4B-stage1c-hrm-fastcap-repeat2
+seed: 0
+skip_batches: 0
+total_steps_override: 700000
+weight_decay: 0.1
+weights_only_resume_from_ema: false

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.0.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:72924ffbd36e656de377461dfe587c7c547967e08729832c211c9addf79673f0
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.1.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8c25143c5e464fac0be235456f161d1390835146f92a1c11cef0c9282e104134
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.2.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc57c488324c10daf89b5ff85e5383e5e36234ced1970ac382884809b0ed8ee
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.3.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:ccfbda50c82b1108c417faed88b93ea6defca7f6a6b363ffb8bc915aed651928
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.4.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:d51535d92436943d8c7ad6d12aa8f9ecf514203b8a1a8c8586b4eeccef0509e3
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.5.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab3bf60ced4993cd036078317a1452a51666d63b0119bedbd504f28fd99f881
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.6.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8ec3d09e0215540c146e7acc2036d714974f8287b688679d2cd1e9b3e4ec1f13
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/carry_step_520000.7.pt ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:09bed99ed90f7c044a3841ba1c57daa5ffbb4a2e79828198ef930aa82c1d7d95
+size 1333

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/.metadata ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:c969c0888fd52ba3483047659af884576dde7947875501052fd911ad8ea1b487
+size 983812

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__0_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3ff2bbabf42d6efeb732e22089306726d8c28c0a96ce05667679b955cbaa0d2a
+size 2769065329

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__1_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:15f45d89a6a25b34997f54f1ea851541ec66d8c4b45fcf278aa1142e980b2a24
+size 2769090643

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__2_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:bb38ebc2f5d67cb8d237334a08344cb0ac6eff188a56a8057d9b31fae7b2c2bb
+size 2769090643

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__3_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:6ee89164b9d3b03f02839f59ea7ad3df83994618aba610610d00126e6c406077
+size 2769090643

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__4_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0d00c71509016e3f94a042d30a68ac0bc218598dfa2a74c374b90580b416ac62
+size 2769090643

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__5_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b44826a5f8fc2adf2476d287a0a81a2239eb814e59be4da1b0daa6e92b81c625
+size 2769090643

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__6_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e6d8379fe00540dac200e609bc80adfbf80cd46b996a0f7a27bd30b1aaf6a032
+size 2769091588

stage1c-hrm-fastcap-repeat2-step520000/fsdp2_step_520000/__7_0.distcp ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:8267c9cecffcc3bea66068db14c6a9cef202b77f0ab683b74744faf6b0816b93
+size 2769098756

stage1c-hrm-fastcap-repeat2-step520000/latest_checkpoint.txt ADDED Viewed

	@@ -0,0 +1 @@


1	+ step_520000

stage1c-hrm-fastcap-repeat2-step520000/step_520000_info.json ADDED Viewed

	@@ -0,0 +1,8 @@

+{
+  "tag": "step_520000",
+  "global_step": 520000,
+  "stage_start_step": 470077,
+  "skip_batches_hint": 49923,
+  "data_path": "/home/work/.data/hrm_text_prepared/koterm_hrm_cleaned_fastcap_stage1_v1",
+  "global_batch_size": 180224
+}

stage1c-hrm-fastcap-repeat2-step520000/train_metadata.yaml ADDED Viewed

	@@ -0,0 +1,13 @@

+max_seq_len: 4096
+tokenizer_info:
+  boq: <|im_start|>
+  condition_mapping:
+    cot: <|object_ref_end|>
+    direct: <|object_ref_start|>
+    noisy: <|quad_start|>
+    synth: <|quad_end|>
+  eoa: <|box_end|>
+  eoq: <|im_end|>
+  tokenizer_path: /home/work/.data/huggingface/trained_tokenizers/hrm-ko-terminal-131k-v1/tokenizer.json
+total_length: 14554291763
+vocab_size: 131072

stage1c-hrm-fastcap-repeat2-step520000/upload_manifest.json ADDED Viewed

	@@ -0,0 +1 @@


1	+ stage1c-hrm-fastcap-repeat2 step_520000 raw resume checkpoint