Filter by type:
Filter by tags:
](images/xhtoken-wechat.jpg){kind=icon}
spark-x2.5-4b
# Spark-X2.5
[](https://join.slack.com/t/tokenspark/shared_invite/zt-432qf8l2f-5~dLyXv8uETr0P0UuC07nw)
[](https://discord.gg/kTDE2Hg8aw)
[](https://www.youtube.com/@SparkLLM)
[](https://dev.to/sparkllm)
[](https://bsky.app/profile/sparkllm.bsky.social)
[](https://x.com/sparkllm)
[](https://www.zhihu.com/people/zhiikz7qh7m)
[](images/xhtoken-wechat.jpg)
> [!Note]
> This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format.
## Introduction
We are introducing Spark-X2.5-4B and Spark-X2.5-1.7B, two compact, general-purpose language models designed to make capable AI more practical, efficient, and accessible. The models deliver strong performance across a broad range of everyday tasks—including conversation, writing, translation, reasoning, coding, tool use, and agentic workflows—achieving leading results among open-source models of comparable size. Spark-X2.5 combines an efficiency-oriented architecture with native context windows of up to 1M tokens, and support for more than 200 languages.
...
Repository: localaiLicense: apache-2.0
qwen3.5-9b-defiant-fable-mtp
Qwen3.5 9B Defiant Fable is an Apache-2.0 multimodal fine-tune for
reasoning, coding, creative writing, and roleplay. It retains the 256K
context window and vision support of Qwen3.5 while reducing refusals.
This default entry uses the NEO-imatrix Q4_K_M build with multi-token
prediction enabled for faster generation.
Repository: localaiLicense: apache-2.0
qwen3.5-9b-defiant-fable
Qwen3.5 9B Defiant Fable in the plain NEO-imatrix Q4_K_M GGUF format.
This fallback offers the same multimodal reasoning, coding, and creative
capabilities without enabling multi-token prediction.
Repository: localaiLicense: apache-2.0
qwen3.5-9b-defiant-fable-q8-mtp
Qwen3.5 9B Defiant Fable in Q8_0 GGUF format for multimodal reasoning,
coding, and creative writing. Includes the matching BF16 vision projector.
Enables multi-token prediction with the MTP weights.
Repository: localaiLicense: apache-2.0
qwen3.5-9b-defiant-fable-q8
Qwen3.5 9B Defiant Fable in Q8_0 GGUF format for multimodal reasoning,
coding, and creative writing. Includes the matching BF16 vision projector.
Uses ordinary decoding without multi-token prediction.
Repository: localaiLicense: apache-2.0
qwen3.8-27b-cold-fusion-q4-mtp
Qwen3.8 27B Cold Fusion is an Apache-2.0 multimodal fine-tune for
reasoning, coding, creative writing, and roleplay. This entry uses the
publisher's NEO-imatrix Q4_K_M GGUF with multi-token prediction enabled.
It supports vision through the shared BF16 projector and a native 256K
context window.
Repository: localaiLicense: apache-2.0
qwen3.8-27b-cold-fusion-q8-mtp
Qwen3.8 27B Cold Fusion in the higher-quality NEO-imatrix Q8_0 GGUF
format. Multi-token prediction is enabled, and the shared BF16 projector
provides vision support.
Repository: localaiLicense: apache-2.0
gemma-4-12b-qat-hauhaucs-balanced
HauhauCS Balanced is an uncensored Gemma 4 12B model built from
quantization-aware-trained weights for chat, coding, and creative writing.
This Q4_K_M GGUF build includes the vision projector for image input.
Repository: localaiLicense: gemma
gemma-4-12b-qat-hauhaucs-balanced-mtp
HauhauCS Balanced is an uncensored Gemma 4 12B model built from
quantization-aware-trained weights for chat, coding, and creative writing.
This Q4_K_M GGUF build includes the vision projector for image input.
The MTP variant adds the publisher-provided draft head and enables
llama.cpp multi-token-prediction speculative decoding.
Repository: localaiLicense: gemma
gemma-4-31b-ortenzya-q4
Ortenzya is a Gemma 4 31B instruction-tuned model from llmfan46 for
creative writing, translation, and roleplay. This entry uses the Q4_K_M
GGUF with the BF16 vision projector and a 32,768-token context.
Repository: localaiLicense: apache-2.0
gemma-4-31b-ortenzya-q8
Ortenzya is a Gemma 4 31B instruction-tuned model from llmfan46 for
creative writing, translation, and roleplay. This entry uses the Q8_0
GGUF with the BF16 vision projector and a 32,768-token context.
Repository: localaiLicense: apache-2.0
cajal-4b-p2pclaw
CAJAL-4B is a Qwen3.5-based 4B GGUF model specialized for scientific paper generation. It is tuned for academic structure, citation formats, LaTeX, and domain-specific technical writing within the P2PCLAW ecosystem, while remaining suitable for local llama-cpp inference.
Repository: localaiLicense: apache-2.0
aurore-reveil_koto-small-7b-it
Koto-Small-7B-IT is an instruct-tuned version of Koto-Small-7B-PT, which was trained on MiMo-7B-Base for almost a billion tokens of creative-writing data. This model is meant for roleplaying and instruct usecases.
Repository: localaiLicense: mit
qwen3-30b-a3b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-30B-A3B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 30.5B in total and 3.3B activated
Number of Paramaters (Non-Embedding): 29.9B
Number of Layers: 48
Number of Attention Heads (GQA): 32 for Q and 4 for KV
Number of Experts: 128
Number of Activated Experts: 8
Context Length: 32,768 natively and 131,072 tokens with YaRN.
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation.
Repository: localaiLicense: apache-2.0
qwen3-32b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-32B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 32.8B
Number of Paramaters (Non-Embedding): 31.2B
Number of Layers: 64
Number of Attention Heads (GQA): 64 for Q and 8 for KV
Context Length: 32,768 natively and 131,072 tokens with YaRN.
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation.
Repository: localaiLicense: apache-2.0
qwen3-14b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-14B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 14.8B
Number of Paramaters (Non-Embedding): 13.2B
Number of Layers: 40
Number of Attention Heads (GQA): 40 for Q and 8 for KV
Context Length: 32,768 natively and 131,072 tokens with YaRN.
For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub, and Documentation.
Repository: localaiLicense: apache-2.0
qwen3-8b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Model Overview
Qwen3-8B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 8.2B
Number of Paramaters (Non-Embedding): 6.95B
Number of Layers: 36
Number of Attention Heads (GQA): 32 for Q and 8 for KV
Context Length: 32,768 natively and 131,072 tokens with YaRN.
Repository: localaiLicense: apache-2.0
qwen3-4b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-4B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 4.0B
Number of Paramaters (Non-Embedding): 3.6B
Number of Layers: 36
Number of Attention Heads (GQA): 32 for Q and 8 for KV
Context Length: 32,768 natively and 131,072 tokens with YaRN.
Repository: localaiLicense: apache-2.0
qwen3-1.7b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-1.7B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 1.7B
Number of Paramaters (Non-Embedding): 1.4B
Number of Layers: 28
Number of Attention Heads (GQA): 16 for Q and 8 for KV
Context Length: 32,768
Repository: localaiLicense: apache-2.0
qwen3-0.6b
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models. Built upon extensive training, Qwen3 delivers groundbreaking advancements in reasoning, instruction-following, agent capabilities, and multilingual support, with the following key features:
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Qwen3-0.6B has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 0.6B
Number of Paramaters (Non-Embedding): 0.44B
Number of Layers: 28
Number of Attention Heads (GQA): 16 for Q and 8 for KV
Context Length: 32,768
Repository: localaiLicense: apache-2.0
shuttleai_shuttle-3.5
A fine-tuned version of Qwen3 32b, emulating the writing style of Claude 3 models and thoroughly trained on role-playing data.
Uniquely support of seamless switching between thinking mode (for complex logical reasoning, math, and coding) and non-thinking mode (for efficient, general-purpose dialogue) within single model, ensuring optimal performance across various scenarios.
Significantly enhancement in its reasoning capabilities, surpassing previous QwQ (in thinking mode) and Qwen2.5 instruct models (in non-thinking mode) on mathematics, code generation, and commonsense logical reasoning.
Superior human preference alignment, excelling in creative writing, role-playing, multi-turn dialogues, and instruction following, to deliver a more natural, engaging, and immersive conversational experience.
Expertise in agent capabilities, enabling precise integration with external tools in both thinking and unthinking modes and achieving leading performance among open-source models in complex agent-based tasks.
Support of 100+ languages and dialects with strong capabilities for multilingual instruction following and translation.
Shuttle 3.5 has the following features:
Type: Causal Language Models
Training Stage: Pretraining & Post-training
Number of Parameters: 32.8B
Number of Paramaters (Non-Embedding): 31.2B
Number of Layers: 64
Number of Attention Heads (GQA): 64 for Q and 8 for KV
Context Length: 32,768 natively and 131,072 tokens with YaRN.
Repository: localaiLicense: apache-2.0
Page 1