Qwen2.5-Turbo (1M)

Released

Qwen2.5-Turbo (1M) is an AI model released by Qwen on Friday, Nov 15 2024, 57 days after Qwen2.5-Math. It has a 1M token context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider.. Benchmark results (shown below) cover LiveCodeBench, HumanEval, IFEval, MATH, MMLU-Pro, and GSM8K.

Benchmarks

Coding

LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.
37.8%
#71 of 86
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better.
86.6%
#29 of 88

Agentic & tool use

IFEvalAgentic tasks — following verifiable instructions in generated responses
76.3%
#20 of 24

Reasoning & science

MATHReasoning — accuracy on the original competition mathematics test set
81.1%
#10 of 60
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better.
93.8%
#10 of 56

Knowledge work

MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark
64.5%
#52 of 58

Compare Qwen2.5-Turbo (1M) with

Qwen2.5-Turbo (1M)

About

Alibaba announced the million-token version of Qwen2.5-Turbo on November 15, 2024. The API release extended the previous 128K context to handle larger document collections through the Turbo tier.

Qwen's technical report identifies the hosted snapshot as qwen-turbo-2024-11-01 and reports 86.6 on HumanEval and 81.1 on MATH. The hosted rollout preceded the downloadable Qwen2.5-1M checkpoints by about two months.

Frequently asked questions

Qwen2.5-Turbo (1M) was released by Qwen on Friday, Nov 15 2024.

All Qwen releases

42 tracked