commit 9a16ab7266145aa93e1c8a413cadf3595ebaf66c Author: ModelHub XC Date: Wed Apr 29 23:00:52 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: RedHatAI/granite-3.1-2b-base-quantized.w8a8 Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..a6344aa --- /dev/null +++ b/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..dbd6798 --- /dev/null +++ b/README.md @@ -0,0 +1,463 @@ +--- +tags: +- w8a8 +- int8 +- vllm +license: apache-2.0 +license_link: https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md +language: + - en +base_model: ibm-granite/granite-3.1-2b-base +library_name: transformers +--- + +# granite-3.1-2b-base-quantized.w8a8 + +## Model Overview +- **Model Architecture:** granite-3.1-2b-base + - **Input:** Text + - **Output:** Text +- **Model Optimizations:** + - **Weight quantization:** INT8 + - **Activation quantization:** INT8 +- **Release Date:** 1/8/2025 +- **Version:** 1.0 +- **Model Developers:** Neural Magic + +Quantized version of [ibm-granite/granite-3.1-2b-base](https://huggingface.co/ibm-granite/granite-3.1-2b-base). +It achieves an average score of 57.22 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 57.65. + +### Model Optimizations + +This model was obtained by quantizing the weights and activations of [ibm-granite/granite-3.1-2b-base](https://huggingface.co/ibm-granite/granite-3.1-2b-base) to INT8 data type, ready for inference with vLLM >= 0.5.2. +This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%. Only the weights and activations of the linear operators within transformers blocks are quantized. + +## Deployment + +### Use with vLLM + +This model can be deployed efficiently using the [vLLM](https://docs.vllm.ai/en/latest/) backend, as shown in the example below. + +```python +from transformers import AutoTokenizer +from vllm import LLM, SamplingParams + +max_model_len, tp_size = 4096, 1 +model_name = "neuralmagic/granite-3.1-2b-base-quantized.w8a8" +tokenizer = AutoTokenizer.from_pretrained(model_name) +llm = LLM(model=model_name, tensor_parallel_size=tp_size, max_model_len=max_model_len, trust_remote_code=True) +sampling_params = SamplingParams(temperature=0.3, max_tokens=256, stop_token_ids=[tokenizer.eos_token_id]) + +messages_list = [ + [{"role": "user", "content": "Who are you? Please respond in pirate speak!"}], +] + +prompt_token_ids = [tokenizer.apply_chat_template(messages, add_generation_prompt=True) for messages in messages_list] + +outputs = llm.generate(prompt_token_ids=prompt_token_ids, sampling_params=sampling_params) + +generated_text = [output.outputs[0].text for output in outputs] +print(generated_text) +``` + +vLLM also supports OpenAI-compatible serving. See the [documentation](https://docs.vllm.ai/en/latest/) for more details. + +## Creation + +This model was created with [llm-compressor](https://github.com/vllm-project/llm-compressor) by running the code snippet below. + +
+ Model Creation Code + +```bash +python quantize.py --model_path ibm-granite/granite-3.1-2b-base --quant_path "output_dir/granite-3.1-2b-base-quantized.w8a8" --calib_size 1024 --dampening_frac 0.01 --observer mse +``` + + +```python +from datasets import load_dataset +from transformers import AutoTokenizer +from llmcompressor.modifiers.quantization import GPTQModifier +from llmcompressor.modifiers.smoothquant import SmoothQuantModifier +from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot, apply +import argparse +from compressed_tensors.quantization import QuantizationScheme, QuantizationArgs, QuantizationType, QuantizationStrategy + + +parser = argparse.ArgumentParser() +parser.add_argument('--model_path', type=str) +parser.add_argument('--quant_path', type=str) +parser.add_argument('--calib_size', type=int, default=256) +parser.add_argument('--dampening_frac', type=float, default=0.1) +parser.add_argument('--observer', type=str, default="minmax") +args = parser.parse_args() + +model = SparseAutoModelForCausalLM.from_pretrained( + args.model_path, + device_map="auto", + torch_dtype="auto", + use_cache=False, + trust_remote_code=True, +) +tokenizer = AutoTokenizer.from_pretrained(args.model_path) + +NUM_CALIBRATION_SAMPLES = args.calib_size +DATASET_ID = "neuralmagic/LLM_compression_calibration" +DATASET_SPLIT = "train" +ds = load_dataset(DATASET_ID, split=DATASET_SPLIT) +ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES)) + +def preprocess(example): + return {"text": example["text"]} + +ds = ds.map(preprocess) + +def tokenize(sample): + return tokenizer( + sample["text"], + padding=False, + truncation=False, + add_special_tokens=True, + ) + + +ds = ds.map(tokenize, remove_columns=ds.column_names) + +ignore=["lm_head"] +mappings=[ + [["re:.*q_proj", "re:.*k_proj", "re:.*v_proj"], "re:.*input_layernorm"], + [["re:.*gate_proj", "re:.*up_proj"], "re:.*post_attention_layernorm"], + [["re:.*down_proj"], "re:.*up_proj"] +] + +recipe = [ + SmoothQuantModifier(smoothing_strength=0.7, ignore=ignore, mappings=mappings), + GPTQModifier( + targets=["Linear"], + ignore=["lm_head"], + scheme="W8A8", + dampening_frac=args.dampening_frac, + observer=args.observer, + ) +] +oneshot( + model=model, + dataset=ds, + recipe=recipe, + num_calibration_samples=args.calib_size, + max_seq_length=8196, +) + +# Save to disk compressed. +model.save_pretrained(quant_path, save_compressed=True) +tokenizer.save_pretrained(quant_path) +``` +
+ +## Evaluation + +The model was evaluated on OpenLLM Leaderboard [V1](https://huggingface.co/spaces/open-llm-leaderboard-old/open_llm_leaderboard), OpenLLM Leaderboard [V2](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard#/) and on [HumanEval](https://github.com/neuralmagic/evalplus), using the following commands: + +
+Evaluation Commands + +OpenLLM Leaderboard V1: +``` +lm_eval \ + --model vllm \ + --model_args pretrained="neuralmagic/granite-3.1-2b-base-quantized.w8a8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=1,gpu_memory_utilization=0.8,enable_chunked_prefill=True,trust_remote_code=True \ + --tasks openllm \ + --write_out \ + --batch_size auto \ + --output_path output_dir \ + --show_config +``` + +#### HumanEval +##### Generation +``` +python3 codegen/generate.py \ + --model neuralmagic/granite-3.1-2b-base-quantized.w8a8 \ + --bs 16 \ + --temperature 0.2 \ + --n_samples 50 \ + --root "." \ + --dataset humaneval +``` +##### Sanitization +``` +python3 evalplus/sanitize.py \ + humaneval/neuralmagic--granite-3.1-2b-base-quantized.w8a8_vllm_temp_0.2 +``` +##### Evaluation +``` +evalplus.evaluate \ + --dataset humaneval \ + --samples humaneval/neuralmagic--granite-3.1-2b-base-quantized.w8a8_vllm_temp_0.2-sanitized +``` +
+ +### Accuracy + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
CategoryMetricibm-granite/granite-3.1-2b-baseneuralmagic/granite-3.1-2b-base-quantized.w8a8Recovery (%)
OpenLLM V1ARC-Challenge (Acc-Norm, 25-shot)53.7554.01100.48
GSM8K (Strict-Match, 5-shot)47.8446.5597.30
HellaSwag (Acc-Norm, 10-shot)77.9477.94100.00
MMLU (Acc, 5-shot)52.8852.3498.98
TruthfulQA (MC2, 0-shot)39.0438.1297.64
Winogrande (Acc, 5-shot)74.4374.3599.89
Average Score57.6557.2299.26
CodingHumanEval Pass@130.0029.6098.67
+ + + +## Inference Performance + + +This model achieves up to 1.4x speedup in single-stream deployment and up to 1.1x speedup in multi-stream asynchronous deployment, depending on hardware and use-case scenario. +The following performance benchmarks were conducted with [vLLM](https://docs.vllm.ai/en/latest/) version 0.6.6.post1, and [GuideLLM](https://github.com/neuralmagic/guidellm). + +
+Benchmarking Command + +``` +guidellm --model neuralmagic/granite-3.1-2b-base-quantized.w8a8 --target "http://localhost:8000/v1" --data-type emulated --data "prompt_tokens=,generated_tokens=" --max seconds 360 --backend aiohttp_server +``` + +
+ + +### Single-stream performance (measured with vLLM version 0.6.6.post1) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Latency (s)
GPU classModelSpeedupCode Completion
prefill: 256 tokens
decode: 1024 tokens		Docstring Generation
prefill: 768 tokens
decode: 128 tokens		Code Fixing
prefill: 1024 tokens
decode: 1024 tokens		RAG
prefill: 1024 tokens
decode: 128 tokens		Instruction Following
prefill: 256 tokens
decode: 128 tokens		Multi-turn Chat
prefill: 512 tokens
decode: 256 tokens		Large Summarization
prefill: 4096 tokens
decode: 512 tokens
A5000granite-3.1-2b-base10.91.411.01.51.42.86.1
granite-3.1-2b-base-quantized.w8a8
(this model)		1.377.91.08.01.11.02.04.7
granite-3.1-2b-base-quantized.w4a161.945.40.75.50.80.71.43.4
A6000granite-3.1-2b-base9.81.310.01.31.32.65.4
granite-3.1-2b-base-quantized.w8a8
(this model)		1.317.81.07.61.00.91.94.5
granite-3.1-2b-base-quantized.w4a161.875.10.75.20.70.71.33.1
+ + +### Multi-stream asynchronous performance (measured with vLLM version 0.6.6.post1) + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
Maximum Throughput (Queries per Second)
GPU classModelSpeedupCode Completion
prefill: 256 tokens
decode: 1024 tokens		Docstring Generation
prefill: 768 tokens
decode: 128 tokens		Code Fixing
prefill: 1024 tokens
decode: 1024 tokens		RAG
prefill: 1024 tokens
decode: 128 tokens		Instruction Following
prefill: 256 tokens
decode: 128 tokens		Multi-turn Chat
prefill: 512 tokens
decode: 256 tokens		Large Summarization
prefill: 4096 tokens
decode: 512 tokens
A5000granite-3.1-2b-base2.910.21.88.219.39.11.3
granite-3.1-2b-base-quantized.w8a8
(this model)		1.133.112.12.09.622.210.21.4
granite-3.1-2b-base-quantized.w4a160.982.810.01.88.118.69.01.2
A6000granite-3.1-2b-base3.712.42.410.323.611.01.6
granite-3.1-2b-base-quantized.w8a8
(this model)		1.123.614.42.712.028.312.91.7
granite-3.1-2b-base-quantized.w4a160.953.711.42.59.822.110.41.4
diff --git a/config.json b/config.json new file mode 100644 index 0000000..9e47553 --- /dev/null +++ b/config.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64de972a12d170123a7aa4281f051a1b9ae16a45bc2380d605fc334e4254089f +size 1906 diff --git a/configuration.json b/configuration.json new file mode 100644 index 0000000..bbeeda1 --- /dev/null +++ b/configuration.json @@ -0,0 +1 @@ +{"framework": "pytorch", "task": "text-generation", "allow_remote": true} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..614b75d --- /dev/null +++ b/generation_config.json @@ -0,0 +1,7 @@ +{ + "_from_model_config": true, + "bos_token_id": 0, + "eos_token_id": 0, + "pad_token_id": 0, + "transformers_version": "4.47.1" +} diff --git a/merges.txt b/merges.txt new file mode 100644 index 0000000..f9f6899 --- /dev/null +++ b/merges.txt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:303127a244b0078878156c17229f36d11b7a3a3f8e47b7cfdbb304ff46be5030 +size 441810 diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..c26caaa --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f56a1e0d6926e6042eef5f15b88daba17e5c547af4080db1e563561f0f2d4f6f +size 2837639392 diff --git a/recipe.yaml b/recipe.yaml new file mode 100644 index 0000000..2c73d48 --- /dev/null +++ b/recipe.yaml @@ -0,0 +1,21 @@ +quant_stage: + quant_modifiers: + SmoothQuantModifier: + smoothing_strength: 0.7 + mappings: + - - ['re:.*q_proj', 're:.*k_proj', 're:.*v_proj'] + - re:.*input_layernorm + - - ['re:.*gate_proj', 're:.*up_proj'] + - re:.*post_attention_layernorm + - - ['re:.*down_proj'] + - re:.*up_proj + GPTQModifier: + sequential_update: true + dampening_frac: 0.01 + ignore: [lm_head] + config_groups: + group_0: + targets: [Linear] + weights: {num_bits: 8, type: int, symmetric: true, strategy: channel, observer: mse} + input_activations: {num_bits: 8, type: int, symmetric: true, strategy: token, dynamic: true, + observer: memoryless} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..7dec8d5 --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,51 @@ +{ + "additional_special_tokens": [ + "<|endoftext|>", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "", + "" + ], + "bos_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "unk_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..0f72a12 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b05515b3c0fced3086c56484444669b1a99a86a8e843dba0bd7c48ac609e9d6 +size 3475389 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..2047874 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a65886cabc33b82795df74cbcf99e88f4be3e93a72808eb7c006a8ae3d66afb3 +size 4156 diff --git a/vocab.json b/vocab.json new file mode 100644 index 0000000..fdc0b09 --- /dev/null +++ b/vocab.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:20175afb9f164fad4829aca2279f8df7eeff1e2e3f671378aaa287a740aff09f +size 776993