Model Detail
Qwen2.5-Coder-32B-Instruct
▼ 1.8%Qwen2.5-Coder-32B-Instruct is a code generation model with 32B parameters released by Qwen. The model is registered under the text-generation pipeline tag on Hugging Face, and supports text->text inputs, distributed under the permissive apache-2.0 license.
Qwen2.5-Coder-32B-Instruct has been evaluated across multiple task suites. On task-specific evaluations the model scores 9.0% resolved on SWE-Bench. Open-LLM-Leaderboard scoring places it at MMLU-Pro 38, GPQA 13, IFEval 73, BBH 52, giving a sense of how it handles instruction following, reasoning, and graduate-level QA in absolute terms.
Qwen2.5-Coder-32B-Instruct is priced at $0.12/M input tokens and $0.3/M output tokens. Operationally the model offers a 128K-token context window, which matters when sizing it for prompt-heavy or latency-sensitive workloads. At this input rate the model sits in the commodity tier and is suitable for high-volume workloads where per-call cost dominates the decision.
Qwen2.5-Coder-32B-Instruct ships as a Qwen2ForCausalLM / 💬 chat models (RLHF, DPO, IFT, ...) architecture with 32B parameters. The published knowledge cutoff is 2024-06-30, so newer events will not be reflected in zero-shot answers without retrieval. Total weight footprint is approximately 32.8 GB, which is the relevant figure when planning local-inference VRAM. The apache-2.0 license is permissive, allowing commercial deployment and derivative work without per-seat fees, though attribution requirements still apply.
Downloads of Qwen2.5-Coder-32B-Instruct have moved -1.8% over the past 24 hours. That is a slight downtrend, consistent with normal cooling as newer models compete for the same workloads. These numbers are signal, not guarantee — week-over-week download counts on Hugging Face also reflect mirror traffic, CI scrapes, and one-off benchmarking runs.
Qwen2.5-Coder-32B-Instruct is best fit for code completion, repository-scale Q&A, and pair-programming integrations, and high-volume batch jobs where per-call cost dominates the budget. It is a less obvious choice for one-shot generation of security-critical code without review. Treat this as a starting matrix rather than a benchmark verdict — the right deployment usually depends on the specific evaluation suite that mirrors your workload.
System Report for CCL25-Eval Task 5: New Dataset and LoRA-Fine-Tuned Qwen2.5
arXiv:2606.12392v1 Announce Type: cross Abstract: Recently, large language models (LLMs) have achieved promising progress in the fields of classical Chinese translation and the generation of classical poetry. However, domain-specific research on precise translation and affective-semantic understandi
Tuning Qwen2.5-VL to Improve Its Web Interaction Skills
arXiv:2604.09571v1 Announce Type: cross Abstract: Recent advances in vision-language models (VLMs) have sparked growing interest in using them to automate web tasks, yet their feasibility as independent agents that reason and act purely from visual input remains underexplored. We investigate this se
