Instructions to use fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym with Transformers:

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

pipe = pipeline("text-generation", model="fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym

SGLang

How to use fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym 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 "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym with Docker Model Runner:
```
docker model run hf.co/fbaldassarri/internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym
```

internlm_internlm3-8b-instruct-autogptq-int8-gs64-sym / config.json

fbaldassarri

Initial Upload

b712d99 verified over 1 year ago

Raw

History Blame Contribute Delete

1.8 kB

	{
	"architectures": [
	"InternLM3ForCausalLM"
	],
	"attention_dropout": 0.0,
	"auto_map": {
	"AutoConfig": "internlm/internlm3-8b-instruct--configuration_internlm3.InternLM3Config",
	"AutoModel": "internlm/internlm3-8b-instruct--modeling_internlm3.InternLM3Model",
	"AutoModelForCausalLM": "modeling_internlm3.InternLM3ForCausalLM"
	},
	"bias": false,
	"bos_token_id": 1,
	"eos_token_id": 2,
	"head_dim": 128,
	"hidden_act": "silu",
	"hidden_size": 4096,
	"initializer_range": 0.02,
	"intermediate_size": 10240,
	"max_position_embeddings": 32768,
	"model_type": "internlm3",
	"num_attention_heads": 32,
	"num_hidden_layers": 48,
	"num_key_value_heads": 2,
	"pad_token_id": 2,
	"qkv_bias": false,
	"quantization_config": {
	"act_bits": 16,
	"act_data_type": "int",
	"act_dynamic": true,
	"act_group_size": 64,
	"act_sym": true,
	"amp": false,
	"autoround_version": "0.4.7",
	"batch_size": 4,
	"bits": 8,
	"damp_percent": 0.01,
	"data_type": "int",
	"desc_act": false,
	"enable_minmax_tuning": true,
	"enable_norm_bias_tuning": false,
	"enable_quanted_input": true,
	"gradient_accumulate_steps": 1,
	"group_size": 64,
	"iters": 200,
	"low_gpu_mem_usage": false,
	"lr": 0.005,
	"minmax_lr": 0.005,
	"nsamples": 128,
	"quant_method": "gptq",
	"scale_dtype": "torch.float16",
	"seqlen": 512,
	"super_bits": null,
	"super_group_size": null,
	"sym": true,
	"to_quant_block_names": null,
	"true_sequential": false
	},
	"rms_norm_eps": 1e-05,
	"rope_scaling": {
	"factor": 6.0,
	"rope_type": "dynamic"
	},
	"rope_theta": 50000000,
	"tie_word_embeddings": false,
	"torch_dtype": "float16",
	"transformers_version": "4.51.2",
	"use_cache": true,
	"vocab_size": 128512
	}