commit bf7865547040da0e7115b340b7fd17381f9f0a06 Author: ModelHub XC Date: Fri Sep 18 07:32:16 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: gabriellarson/LFM2-8B-A1B-GGUF Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..fb134f8 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,57 @@ +*.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 +LFM2-8B-A1B-F16.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ4_NL.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q5_0.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ3_M.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ3_S.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ4_XS.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-MXFP4_MOE.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ2_M.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ2_S.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ2_XXS.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-IQ3_XXS.gguf filter=lfs diff=lfs merge=lfs -text +LFM2-8B-A1B-Q2_K.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/LFM2-8B-A1B-F16.gguf b/LFM2-8B-A1B-F16.gguf new file mode 100644 index 0000000..19bd1ff --- /dev/null +++ b/LFM2-8B-A1B-F16.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1eaf4f0d5cb6cf5f897cc3a5e14d9de88c85eb0393ba1ad6f85480bc3b63445e +size 16685562560 diff --git a/LFM2-8B-A1B-IQ2_M.gguf b/LFM2-8B-A1B-IQ2_M.gguf new file mode 100644 index 0000000..7e3f318 --- /dev/null +++ b/LFM2-8B-A1B-IQ2_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f97531f68199cb5ac8fc2d5eee07160c0b63f56fae8248c5d58f82e7959d046d +size 2749479840 diff --git a/LFM2-8B-A1B-IQ2_S.gguf b/LFM2-8B-A1B-IQ2_S.gguf new file mode 100644 index 0000000..2160ed7 --- /dev/null +++ b/LFM2-8B-A1B-IQ2_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b506119acb2d73c9ee6ac18fd850246572d151ef81b2c9a9b8d6e5847e541e9a +size 2498673568 diff --git a/LFM2-8B-A1B-IQ2_XXS.gguf b/LFM2-8B-A1B-IQ2_XXS.gguf new file mode 100644 index 0000000..90ce0ba --- /dev/null +++ b/LFM2-8B-A1B-IQ2_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:640393f1c11e61d2514827389a871f7d7f5594e81504f63dbb131d6c4884f573 +size 2228632480 diff --git a/LFM2-8B-A1B-IQ3_M.gguf b/LFM2-8B-A1B-IQ3_M.gguf new file mode 100644 index 0000000..d7308d1 --- /dev/null +++ b/LFM2-8B-A1B-IQ3_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fe56618c4cf12d889cf5510ce152068af96f5b4f52769a1fb04fef5f34b56c2e +size 3667966880 diff --git a/LFM2-8B-A1B-IQ3_S.gguf b/LFM2-8B-A1B-IQ3_S.gguf new file mode 100644 index 0000000..88494bd --- /dev/null +++ b/LFM2-8B-A1B-IQ3_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1b090ccd088ff16bc92229602a8c56a7dc3f5377c4237f9cbe295fd5a8e1c8b +size 3644292000 diff --git a/LFM2-8B-A1B-IQ3_XXS.gguf b/LFM2-8B-A1B-IQ3_XXS.gguf new file mode 100644 index 0000000..5858490 --- /dev/null +++ b/LFM2-8B-A1B-IQ3_XXS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:572950d387f66e4fbce62b49ce7b944b0b56222217d8a104b10d6f7d2640774f +size 3241114528 diff --git a/LFM2-8B-A1B-IQ4_NL.gguf b/LFM2-8B-A1B-IQ4_NL.gguf new file mode 100644 index 0000000..0743987 --- /dev/null +++ b/LFM2-8B-A1B-IQ4_NL.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2c0b7439201036ace0122399bbe7ec4632323394d7d47f5ca0ac2aafe3a22761 +size 4733893536 diff --git a/LFM2-8B-A1B-IQ4_XS.gguf b/LFM2-8B-A1B-IQ4_XS.gguf new file mode 100644 index 0000000..92bedb6 --- /dev/null +++ b/LFM2-8B-A1B-IQ4_XS.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:146aaacc6a83c0d1e13450cd12bf3a3b2bc70543cfa7531cf4fedd50c6ab6627 +size 4477516704 diff --git a/LFM2-8B-A1B-MXFP4_MOE.gguf b/LFM2-8B-A1B-MXFP4_MOE.gguf new file mode 100644 index 0000000..83a9e8d --- /dev/null +++ b/LFM2-8B-A1B-MXFP4_MOE.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1019869c521955beaf09b965b88cd2ee67444a8fcc233fd8d35a380de3c5acd0 +size 4750670752 diff --git a/LFM2-8B-A1B-Q2_K.gguf b/LFM2-8B-A1B-Q2_K.gguf new file mode 100644 index 0000000..8a8e779 --- /dev/null +++ b/LFM2-8B-A1B-Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1b1dae52ced526404415cb9db2ee9d4b57a7c5514ec28f1f55f35202556e5fef +size 3079650208 diff --git a/LFM2-8B-A1B-Q3_K_L.gguf b/LFM2-8B-A1B-Q3_K_L.gguf new file mode 100644 index 0000000..ba3c767 --- /dev/null +++ b/LFM2-8B-A1B-Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a9d6728e97d018f58e877a8dc9baa3829d6aeac52cfc7e82591d0e8780b503e8 +size 4326079392 diff --git a/LFM2-8B-A1B-Q3_K_M.gguf b/LFM2-8B-A1B-Q3_K_M.gguf new file mode 100644 index 0000000..1fefb69 --- /dev/null +++ b/LFM2-8B-A1B-Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d92059be7e5e1426caa8bc3c0a5217a6ef17166b529ff638f3d8f9dd4a9122ea +size 3997612960 diff --git a/LFM2-8B-A1B-Q3_K_S.gguf b/LFM2-8B-A1B-Q3_K_S.gguf new file mode 100644 index 0000000..7f3fdae --- /dev/null +++ b/LFM2-8B-A1B-Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6bbdbfd2aca7061a60d932245239b075f7b07818adbf3d28cc30719ebc26f2f4 +size 3644292000 diff --git a/LFM2-8B-A1B-Q4_0.gguf b/LFM2-8B-A1B-Q4_0.gguf new file mode 100644 index 0000000..d1dab3f --- /dev/null +++ b/LFM2-8B-A1B-Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dce55a20b8f904fb3cb8e88dc995478608702345cff62ad1ec86169b1938bd61 +size 4743068576 diff --git a/LFM2-8B-A1B-Q4_K_M.gguf b/LFM2-8B-A1B-Q4_K_M.gguf new file mode 100644 index 0000000..dccad97 --- /dev/null +++ b/LFM2-8B-A1B-Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e4698a5098897313da3141737b627486b59ef8c5ce3916d1f82c8cb0a94c3fc4 +size 5044779936 diff --git a/LFM2-8B-A1B-Q4_K_S.gguf b/LFM2-8B-A1B-Q4_K_S.gguf new file mode 100644 index 0000000..d6199e7 --- /dev/null +++ b/LFM2-8B-A1B-Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09a486859775c3b4b11f6d03a5ca3e7db460bd441a3cefc823a1836a28096ff2 +size 4752767904 diff --git a/LFM2-8B-A1B-Q5_0.gguf b/LFM2-8B-A1B-Q5_0.gguf new file mode 100644 index 0000000..bd47696 --- /dev/null +++ b/LFM2-8B-A1B-Q5_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:51dc44a3fa268c2d9e15d5aae1a399f0b011028b4906f28a1fe45938fe6d5e11 +size 5768575904 diff --git a/LFM2-8B-A1B-Q5_K_M.gguf b/LFM2-8B-A1B-Q5_K_M.gguf new file mode 100644 index 0000000..bb9c610 --- /dev/null +++ b/LFM2-8B-A1B-Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5805d5e928c610d45bf77468e03c54b81483266f374a4d6fc386619fc386c30c +size 5919554464 diff --git a/LFM2-8B-A1B-Q5_K_S.gguf b/LFM2-8B-A1B-Q5_K_S.gguf new file mode 100644 index 0000000..cd3625d --- /dev/null +++ b/LFM2-8B-A1B-Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e840fcb8d39a11012006a1cd468ddb65876e6f568f7c639968b1c11ec0f8aff0 +size 5759400864 diff --git a/LFM2-8B-A1B-Q6_K.gguf b/LFM2-8B-A1B-Q6_K.gguf new file mode 100644 index 0000000..7a6b7c0 --- /dev/null +++ b/LFM2-8B-A1B-Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0a1f3b69218625c6f490004d63ac071daa7e5874d3e7f15b5c05d7a3e588a816 +size 6849002400 diff --git a/LFM2-8B-A1B-Q8_0.gguf b/LFM2-8B-A1B-Q8_0.gguf new file mode 100644 index 0000000..53cbfc3 --- /dev/null +++ b/LFM2-8B-A1B-Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0741ded1af2259b9015a5c55baca1ec06c4b8c3a6f3f75660c490ab78690795c +size 8868428704 diff --git a/README.md b/README.md new file mode 100644 index 0000000..0ccec2c --- /dev/null +++ b/README.md @@ -0,0 +1,344 @@ +--- +library_name: transformers +license: other +license_name: lfm1.0 +license_link: LICENSE +language: +- en +- ar +- zh +- fr +- de +- ja +- ko +- es +pipeline_tag: text-generation +tags: +- liquid +- lfm2 +- edge +- moe +base_model: +- LiquidAI/LFM2-8B-A1B +--- +
+
+ Liquid AI +
+
+ + + Playground + + + + + + + + + + + + + + + + + + + Leap + + + + + + + + + + + + + + +
+
+# LFM2-8B-A1B + +LFM2 is a new generation of hybrid models developed by [Liquid AI](https://www.liquid.ai/blog/lfm2-8b-a1b-an-efficient-on-device-mixture-of-experts), specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency. + +We're releasing the weights of our first MoE based on LFM2, with 8.3B total parameters and 1.5B active parameters. + +- LFM2-8B-A1B is the best on-device MoE in terms of both **quality** (comparable to 3-4B dense models) and **speed** (faster than Qwen3-1.7B). +- **Code and knowledge** capabilities are significantly improved compared to LFM2-2.6B. +- Quantized variants fit comfortably on high-end **phones, tablets, and laptops**. + +Find more information about LFM2-8B-A1B in our [blog post](https://www.liquid.ai/blog/). + +## 📄 Model details + +Due to their small size, **we recommend fine-tuning LFM2 models on narrow use cases** to maximize performance. +They are particularly suited for agentic tasks, data extraction, RAG, creative writing, and multi-turn conversations. +However, we do not recommend using them for tasks that are knowledge-intensive or require programming skills. + +| Property | [**LFM2-8B-A1B**](https://huggingface.co/LiquidAI/LFM2-8B-A1B) | +| --------------------- | ----------------------------- | +| **Total parameters** | 8.3B | +| **Active parameters** | 1.5B | +| **Layers** | 24 (10 conv + 6 attn) | +| **Context length** | 32,768 tokens | +| **Vocabulary size** | 65,536 | +| **Training precision**| Mixed BF16/FP8 | +| **Training budget** | 12 trillion tokens | +| **License** | LFM Open License v1.0 | + +**Supported languages**: English, Arabic, Chinese, French, German, Japanese, Korean, and Spanish. + +**Generation parameters**: We recommend the following parameters: +* `temperature=0.3` +* `min_p=0.15` +* `repetition_penalty=1.05` + +**Chat template**: LFM2 uses a ChatML-like chat template as follows: + +``` +<|startoftext|><|im_start|>system +You are a helpful assistant trained by Liquid AI.<|im_end|> +<|im_start|>user +What is C. elegans?<|im_end|> +<|im_start|>assistant +It's a tiny nematode that lives in temperate soil environments.<|im_end|> +``` + +You can automatically apply it using the dedicated [`.apply_chat_template()`](https://huggingface.co/docs/transformers/en/chat_templating#applychattemplate) function from Hugging Face transformers. + +**Tool use**: It consists of four main steps: +1. **Function definition**: LFM2 takes JSON function definitions as input (JSON objects between `<|tool_list_start|>` and `<|tool_list_end|>` special tokens), usually in the system prompt +2. **Function call**: LFM2 writes Pythonic function calls (a Python list between `<|tool_call_start|>` and `<|tool_call_end|>` special tokens), as the assistant answer. +3. **Function execution**: The function call is executed and the result is returned (string between `<|tool_response_start|>` and `<|tool_response_end|>` special tokens), as a "tool" role. +4. **Final answer**: LFM2 interprets the outcome of the function call to address the original user prompt in plain text. + +Here is a simple example of a conversation using tool use: + +``` +<|startoftext|><|im_start|>system +List of tools: <|tool_list_start|>[{"name": "get_candidate_status", "description": "Retrieves the current status of a candidate in the recruitment process", "parameters": {"type": "object", "properties": {"candidate_id": {"type": "string", "description": "Unique identifier for the candidate"}}, "required": ["candidate_id"]}}]<|tool_list_end|><|im_end|> +<|im_start|>user +What is the current status of candidate ID 12345?<|im_end|> +<|im_start|>assistant +<|tool_call_start|>[get_candidate_status(candidate_id="12345")]<|tool_call_end|>Checking the current status of candidate ID 12345.<|im_end|> +<|im_start|>tool +<|tool_response_start|>{"candidate_id": "12345", "status": "Interview Scheduled", "position": "Clinical Research Associate", "date": "2023-11-20"}<|tool_response_end|><|im_end|> +<|im_start|>assistant +The candidate with ID 12345 is currently in the "Interview Scheduled" stage for the position of Clinical Research Associate, with an interview date set for 2023-11-20.<|im_end|> +``` + +**Architecture**: Hybrid model with multiplicative gates and short convolutions: 18 double-gated short-range LIV convolution blocks and 6 grouped query attention (GQA) blocks. + +**Pre-training mixture**: Approximately 75% English, 20% multilingual, and 5% code data sourced from the web and licensed materials. + +**Training approach**: +* Very large-scale SFT on 50% downstream tasks, 50% general domains +* Custom DPO with length normalization and semi-online datasets +* Iterative model merging + +## 🏃 How to run LFM2 + +### 1. Transformers + +To run LFM2, you need to install Hugging Face [`transformers`](https://github.com/huggingface/transformers) from source as follows: +```bash +pip install git+https://github.com/huggingface/transformers.git@0c9a72e4576fe4c84077f066e585129c97bfd4e6 +``` + +Here is an example of how to generate an answer with transformers in Python: + +```python +from transformers import AutoModelForCausalLM, AutoTokenizer +# Load model and tokenizer +model_id = "LiquidAI/LFM2-8B-A1B" +model = AutoModelForCausalLM.from_pretrained( + model_id, + device_map="auto", + dtype="bfloat16", +# attn_implementation="flash_attention_2" <- uncomment on compatible GPU +) +tokenizer = AutoTokenizer.from_pretrained(model_id) +# Generate answer +prompt = "What is C. elegans?" +input_ids = tokenizer.apply_chat_template( + [{"role": "user", "content": prompt}], + add_generation_prompt=True, + return_tensors="pt", + tokenize=True, +).to(model.device) +output = model.generate( + input_ids, + do_sample=True, + temperature=0.3, + min_p=0.15, + repetition_penalty=1.05, + max_new_tokens=512, +) +print(tokenizer.decode(output[0], skip_special_tokens=False)) +# <|startoftext|><|im_start|>user +# What is C. elegans?<|im_end|> +# <|im_start|>assistant +# C. elegans, also known as Caenorhabditis elegans, is a small, free-living +# nematode worm (roundworm) that belongs to the phylum Nematoda. +``` + +You can directly run and test the model with this [Colab notebook](https://colab.research.google.com/drive/1i0u7X6qen9UJkV6xSCDZ0NmPUH50SOvO?usp=sharing). + +### 2. vLLM + +You can run the model in [`vLLM`](https://github.com/vllm-project/vllm) by building from source: + +```bash +git clone https://github.com/vllm-project/vllm.git +cd vllm +pip install -e . -v +``` + +Here is an example of how to use it for inference: + +```python +from vllm import LLM, SamplingParams +prompts = [ + [ + { + "content": "What is C. elegans?", + "role": "user", + }, + ], + [ + { + "content": "Say hi in JSON format", + "role": "user", + }, + ], + [ + { + "content": "Define AI in Spanish", + "role": "user", + }, + ], +] +sampling_params = SamplingParams( + temperature=0.3, + min_p=0.15, + repetition_penalty=1.05, + max_tokens=30 +) +llm = LLM(model="LiquidAI/LFM2-8B-A1B", dtype="bfloat16") +outputs = llm.chat(prompts, sampling_params) +for i, output in enumerate(outputs): + prompt = prompts[i][0]["content"] + generated_text = output.outputs[0].text + print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}") +``` + + +### 3. llama.cpp + +You can run LFM2 with llama.cpp using its [GGUF checkpoint](https://huggingface.co/LiquidAI/LFM2-8B-A1B-GGUF). Find more information in the model card. + +## 🔧 How to fine-tune LFM2 + +We recommend fine-tuning LFM2 models on your use cases to maximize performance. + +| Notebook | Description | Link | +|-------|------|------| +| SFT (TRL) | Supervised Fine-Tuning (SFT) notebook with a LoRA adapter using TRL. | Colab link | +| DPO (TRL) | Preference alignment with Direct Preference Optimization (DPO) using TRL. | Colab link | + +## 📈 Performance + +### 1. Automated benchmarks + +
+
+ + Benchmarks + +
+
+Compared to similar-sized models, LFM2-8B-A1B displays strong performance in instruction following and math while also running significantly faster. + +| Model | MMLU | MMLU-Pro | GPQA | IFEval | IFBench | Multi-IF | +|---|---|---|---|---|---|---| +| **LFM2-8B-A1B** | 64.84 | 37.42 | 29.29 | 77.58 | 25.85 | 58.19 | +| LFM2-2.6B | 64.42 | 25.96 | 26.57 | 79.56 | 22.19 | 60.26 | +| Llama-3.2-3B-Instruct | 60.35 | 22.25 | 30.6 | 71.43 | 20.78 | 50.91 | +| SmolLM3-3B | 59.84 | 23.90 | 26.31 | 72.44 | 17.93 | 58.86 | +| gemma-3-4b-it | 58.35 | 34.76 | 29.51 | 76.85 | 23.53 | 66.61 | +| Qwen3-4B-Instruct-2507 | 72.25 | 52.31 | 34.85 | 85.62 | 30.28 | 75.54 | +| granite-4.0-h-tiny | 66.79 | 32.03 | 26.46 | 81.06 | 18.37 | 52.99 | + +| Model | GSM8K | GSMPlus | MATH 500 | MATH Lvl 5 | MGSM | MMMLU | +|---|---|---|---|---|---|---| +| **LFM2-8B-A1B** | 84.38 | 64.76 | 74.2 | 62.38 | 72.4 | 55.26 | +| LFM2-2.6B | 82.41 | 60.75 | 63.6 | 54.38 | 74.32 | 55.39 | +| Llama-3.2-3B-Instruct | 75.21 | 38.68 | 41.2 | 24.06 | 61.68 | 47.92 | +| SmolLM3-3B | 81.12 | 58.91 | 73.6 | 51.93 | 68.72 | 50.02 | +| gemma-3-4b-it | 89.92 | 68.38 | 73.2 | 52.18 | 87.28 | 50.14 | +| Qwen3-4B-Instruct-2507 | 68.46 | 56.16 | 85.6 | 73.62 | 81.76 | 60.67 | +| granite-4.0-h-tiny | 82.64 | 59.14 | 58.2 | 36.11 | 73.68 | 56.13 | + +| Model | Active params | LCB v6 | LCB v5 | HumanEval+ | Creative Writing v3 | +|----------------------------|---------------|---------------|---------------|--------------------|-----------------------------| +| **LFM2-8B-A1B** | 1.5B | 21.04% | 21.36% | 69.51% | 44.22% | +| Gemma-3-1b-it | 1B | 4.27% | 4.43% | 37.20% | 41.67% | +| Granite-4.0-h-tiny | 1B | 26.73% | 27.27% | 73.78% | 32.60% | +| Llama-3.2-1B-Instruct | 1.2B | 4.08% | 3.64% | 23.17% | 31.43% | +| Qwen2.5-1.5B-Instruct | 1.5B | 11.18% | 10.57% | 48.78% | 22.18% | +| Qwen3-1.7B (/no_think) | 1.7B | 24.07% | 26.48% | 60.98% | 31.56% | +| LFM2-2.6B | 2.6B | 14.41% | 14.43% | 57.93% | 38.79% | +| SmolLM3-3B | 3.1B | 19.05% | 19.20% | 60.37% | 36.44% | +| Llama-3.2-3B-Instruct | 3.2B | 11.47% | 11.48% | 24.06% | 38.84% | +| Qwen3-4B (/no_think) | 4B | 36.11% | 38.64% | 71.95% | 37.49% | +| Qwen3-4B-Instruct-2507 | 4B | 48.72% | 50.80% | 82.32% | 51.71% | +| Gemma-3-4b-it | 4.3B | 18.86% | 19.09% | 62.8% | 68.56% | + +### 2. Inference + +LFM2-8B-A1B is significantly faster than models with a similar number of active parameters, like Qwen3-1.7B. + +
+
+ + Decode Throughput - S24 Ultra + +
+ +
+ + Decode Throughput - HX370 + +
+
+The following plots showcase the performance of different models under int4 quantization with int8 dynamic activations on the AMD Ryzen AI 9 HX 370 CPU, using 16 threads. The results are obtained using our internal XNNPACK-based inference stack, and a custom CPU MoE kernel. + +
+
+ + Prefill Throughput vs Sequence Length + +
+
+ + Decode Throughput vs Sequence Length + +
+
+## 📬 Contact + +If you are interested in custom solutions with edge deployment, please contact [our sales team](https://www.liquid.ai/contact). \ No newline at end of file