Instructions to use ibm-ai-platform/Bamba-9B-fp8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ibm-ai-platform/Bamba-9B-fp8 with Transformers:

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

pipe = pipeline("text-generation", model="ibm-ai-platform/Bamba-9B-fp8")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ibm-ai-platform/Bamba-9B-fp8")
model = AutoModelForCausalLM.from_pretrained("ibm-ai-platform/Bamba-9B-fp8")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ibm-ai-platform/Bamba-9B-fp8 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ibm-ai-platform/Bamba-9B-fp8"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ibm-ai-platform/Bamba-9B-fp8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ibm-ai-platform/Bamba-9B-fp8

SGLang

How to use ibm-ai-platform/Bamba-9B-fp8 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 "ibm-ai-platform/Bamba-9B-fp8" \
    --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": "ibm-ai-platform/Bamba-9B-fp8",
		"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 "ibm-ai-platform/Bamba-9B-fp8" \
        --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": "ibm-ai-platform/Bamba-9B-fp8",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ibm-ai-platform/Bamba-9B-fp8 with Docker Model Runner:
```
docker model run hf.co/ibm-ai-platform/Bamba-9B-fp8
```

Bamba-9B-fp8 / config.json

divykum

weights upload

caf4659 over 1 year ago

raw

history blame contribute delete

2.17 kB

	{
	"_name_or_path": "/net/storage149/autofs/css22/nwang/118/nvme1/nwang/hf_log/cache/hub/models--ibm-fms--Bamba-9.8b-2.2T-hf/snapshots/d5f0d5c305310e8eec0e727fe3263b9928d6f0b1",
	"architectures": [
	"BambaForCausalLM"
	],
	"attention_dropout": 0.0,
	"attn_layer_indices": [
	9,
	18,
	27
	],
	"attn_rotary_emb": 64,
	"bos_token_id": 128000,
	"eos_token_id": 128001,
	"hidden_act": "silu",
	"hidden_size": 4096,
	"initializer_range": 0.02,
	"intermediate_size": 14336,
	"mamba_chunk_size": 256,
	"mamba_conv_bias": true,
	"mamba_d_conv": 4,
	"mamba_d_head": 64,
	"mamba_d_state": 128,
	"mamba_dt_rank": 256,
	"mamba_expand": 2,
	"mamba_n_groups": 1,
	"mamba_n_heads": 128,
	"mamba_proj_bias": false,
	"model_type": "bamba",
	"num_attention_heads": 32,
	"num_hidden_layers": 32,
	"num_key_value_heads": 8,
	"num_logits_to_keep": 1,
	"pad_token_id": 0,
	"quantization_config": {
	"config_groups": {
	"group_0": {
	"input_activations": {
	"actorder": null,
	"block_structure": null,
	"dynamic": true,
	"group_size": null,
	"num_bits": 8,
	"observer": null,
	"observer_kwargs": {},
	"strategy": "tensor",
	"symmetric": true,
	"type": "float"
	},
	"output_activations": null,
	"targets": [
	"Linear"
	],
	"weights": {
	"actorder": null,
	"block_structure": null,
	"dynamic": false,
	"group_size": null,
	"num_bits": 8,
	"observer": "minmax",
	"observer_kwargs": {},
	"strategy": "tensor",
	"symmetric": true,
	"type": "float"
	}
	}
	},
	"format": "float-quantized",
	"global_compression_ratio": 1.3051470554436795,
	"ignore": [
	"lm_head"
	],
	"kv_cache_scheme": null,
	"quant_method": "compressed-tensors",
	"quantization_status": "compressed"
	},
	"rms_norm_eps": 1e-05,
	"tie_word_embeddings": false,
	"torch_dtype": "bfloat16",
	"transformers_version": "4.47.0.dev0",
	"use_cache": true,
	"use_mamba_kernels": true,
	"vocab_size": 128256
	}