51

Qwen2.5-1M Release: Open‑Source 7B and 14B Models with 1M‑Token Context and vLLM‑Based Inference Framework

Qwen releases open‑source Qwen2.5‑7B‑Instruct‑1M and Qwen2.5‑14B‑Instruct‑1M models that support up to 1 million token contexts, accompanied by an optimized vLLM‑based inference framework that delivers 3×–7× faster prefill and retains short‑task performance comparable to GPT‑4o‑mini.

52

Qwen2.5-VL Release Notes

Qwen has released Qwen2.5-VL, a flagship vision-language model available in 3B, 7B, and 72B sizes that introduces advanced visual agent capabilities, long-video comprehension, and structured document parsing.

53

Qwen Global-Batch Load Balance for MoE LLM Training

Qwen introduces a global-batch load balancing loss that improves MoE model performance and enables expert specialization by calculating balance across the entire global batch rather than individual micro-batches.

54

Qwen2.5-Math-PRM and ProcessBench Release

Qwen has released Qwen2.5-Math-PRM-7B and 72B, state-of-the-art Process Reward Models designed to identify intermediate reasoning errors in mathematical problem solving, alongside ProcessBench, a new step-level evaluation benchmark.

55

QVQ-72B-Preview Release

Qwen has released QVQ-72B-Preview, an open-weight multimodal reasoning model based on Qwen2-VL-72B that achieves a score of 70.3 on the MMMU benchmark.

56

QwQ-32B-Preview: Exploring Deep Reasoning Capabilities

Qwen has introduced QwQ-32B-Preview, a model designed for deep reasoning and complex problem-solving through an internal chain-of-thought process.

57

Qwen2.5-Turbo 1M Token Context Length Release

Qwen has released Qwen2.5-Turbo, which extends the model's context window to 1 million tokens while significantly improving inference speed and maintaining competitive performance on short-sequence tasks.

58

Qwen2.5-Coder Series Release Notes

Qwen has released the Qwen2.5-Coder series, featuring six model sizes from 0.5B to 32B, with the 32B-Instruct model achieving SOTA open-source performance comparable to GPT-4o in coding tasks.

59

Qwen2.5 Release Notes: New Foundation, Coder, and Math Models

Qwen has released Qwen2.5, a comprehensive suite of open-weight dense decoder-only models including general-purpose LLMs, specialized Coder and Math variants, and an updated Qwen2-VL-72B.

60

Qwen2.5 LLM series release

Qwen announced the Qwen2.5 series – open-source decoder-only LLMs from 0.5B to 72B parameters that double the capability of Qwen2 while adding a larger 18‑trillion‑token dataset, major gains in knowledge, coding, math, and alignment, and a 128K token context window.

61

Qwen2.5-Coder Release Notes

Qwen has released Qwen2.5-Coder, a series of open-source coding models trained on 5.5 trillion tokens that outperform larger models in code generation, reasoning, and mathematics.

62

Qwen2.5-Math Release Notes

Qwen has released Qwen2.5-Math, a series of open-source mathematical LLMs supporting bilingual reasoning and Tool-Integrated Reasoning (TIR) to outperform leading closed-source models on complex math benchmarks.

63

Qwen2-VL release: open-source 2B/7B vision-language models and 72B API with state-of-the-art image, video, and multilingual capabilities

Qwen released Qwen2-VL, a new vision-language model series (2B, 7B open-source, 72B API) that sets state-of-the-art performance on image, document, multilingual, and long-video understanding while supporting visual agent capabilities.

64

Qwen2-Audio Release Notes

Qwen2-Audio is an audio-language model capable of voice chat and audio analysis across more than eight languages and dialects, surpassing previous state-of-the-art performance on multiple benchmarks.

65

Qwen2-Math Release Notes

Qwen has released Qwen2-Math, a series of specialized mathematical LLMs (1.5B, 7B, and 72B) that outperform several closed-source models, including GPT-4o, on math benchmarks.

66

Qwen2 Release Notes / What's New

Qwen announces Qwen2, a series of five open-source models ranging from 0.5B to 72B parameters with enhanced multilingual support for 27 additional languages and context lengths up to 128K tokens.

67

Qwen-Agent: Generalizing LLMs from 8k to 1M Context

Qwen has developed an agent framework that enables 8k-context models to process 1M tokens, surpassing both RAG and native long-context models in specific benchmarks.

68

Qwen-Max-0428 Release Notes

Qwen has released Qwen-Max-0428, an instruction-tuned chat model that outperforms Qwen1.5-110B-Chat on MT-Bench and ranks in the top 10 of the Chatbot Arena leaderboard.

69

Qwen1.5-110B release notes / what's new

Qwen has released Qwen1.5-110B, the first model in the Qwen1.5 series to exceed 100 billion parameters, delivering competitive performance against Llama-3-70B and significant improvements over Qwen1.5-72B.

70

CodeQwen1.5 Release Notes

Qwen has released CodeQwen1.5-7B, an open-source code LLM supporting 92 programming languages and 64K token context windows to enhance developer productivity.

71

Qwen1.5-32B release notes / what's new

Qwen has released Qwen1.5-32B and Qwen1.5-32B-Chat, models designed to balance high performance with lower memory and inference costs compared to the 72B version.

72

Qwen1.5-MoE-A2.7B Release Notes

Qwen introduces Qwen1.5-MoE-A2.7B, a Mixture-of-Experts model that matches the performance of 7B dense models while using only 2.7 billion activated parameters.

73

Qwen1.5 Release Notes / What's New

Qwen has released Qwen1.5, a series of open-source base and chat models ranging from 0.5B to 110B parameters, featuring improved human alignment, multilingual capabilities, and native Hugging Face transformers integration.

74

Qwen-VL-Plus and Qwen-VL-Max Release

Qwen has released Qwen-VL-Plus and Qwen-VL-Max, large visual language models that match GPT-4V and Gemini Ultra in multimodal tasks and outperform them in Chinese text comprehension.

75

Introducing Qwen: A Comprehensive LLM and LMM Framework

Qwen is a project towards AGI consisting of a series of multilingual large language models (LLMs) and large multimodal models (LMMs), including open-source versions ranging from 1.8B to 72B parameters.

76

OFASys: A Framework for Multimodal Multitask Learning

Qwen introduces OFASys, an AI framework that simplifies multimodal multitask learning by allowing users to define complex tasks and modalities via a single-line Instruction interface.

77

Chinese CLIP: Contrastive Vision-Language Pretraining in Chinese

Qwen has released Chinese CLIP, a vision-language model designed to overcome the cultural and linguistic limitations of English-centric CLIP models in cross-modal retrieval and image classification.

78

OFA: Towards Building a One-For-All Model

OFA is a unified multimodal pretrained model that unifies understanding and generation tasks across modalities into a single framework using instruction-based multitask pretraining.