Qwen2.5-7B-Instruct
- In Input modalities: Text Out Output modalities: Text
- Tool use
- Structured output
- Open weights
Prices
2 offerings · prices as listedqwen2.5-7b-instruct| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard | ¥0.50 | ¥1.00 |
Prices from How to check model training and deployment pricing-Alibaba Cloud Model Studio(Model Studio)-阿里云帮助中心 (opens in a new tab) (checked Nov 1, 2026)
qwen2.5-7b-instruct| Conditions/ 1M tokens | Input | Output |
|---|---|---|
| Standard no longer listed since Oct 1, 2026 past price, from the vendor’s page archived Jan 16, 2026 | $0.175 | $0.70 |
Prices from Alibaba Cloud Model Studio pricing (opens in a new tab) (checked Sep 30, 2026)
Standard: the price a request with no conditions pays (the headline price elsewhere on the site).
Details
- Reasoning
- —
- Knowledge cutoff
- Apr 2024
- Licence
- apache-2.0
- Family
- qwen
- Model ID
Not stated on Alibaba Qwen’s pages: max input, reasoning efforts, reasoning can be off, max thinking budget, built-in tools, aliases.
Run locally
Weights: Qwen/Qwen2.5-7B-Instruct- Parameters 7.6B
- Dense
- Native precision BF16
- Layers 28
- KV cache per token 56 KB
- Config positions 32K (config.json max_position_embeddings)
- API context 977K (maker’s API)
- Needs
- ≈ 15.3 GB (weights 14.2 (maker’s file) + KV cache 0.4 + overhead 0.7)
- Usable
- 21.6 GB (90% of its memory)
- Generation speed
- 32–48 tok/s
- Longest context
- 32K (the model’s limit)
- First token
- about 0.25–0.37 s for a 2K prompt
| Precision | Weights | Needs at 8K | On your hardware | Longest context |
|---|---|---|---|---|
| BF16 (native)maker’s file | 14.2 GB | ≈ 15.3 | ●Runs well 32–48 tok/s | 32K |
| Q8_0maker’s file | 7.5 GB | ≈ 8.5 | ●Runs well 59–88 tok/s | 32K |
| INT8 (GPTQ)maker’s file | 8.3 GB | ≈ 9.2 | ●Runs well 54–81 tok/s | 32K |
| FP8 | 7.1 GB | ≈ 8.0 | ●Runs well 62–94 tok/s | 32K |
| Q6_Kmaker’s file | 5.8 GB | ≈ 6.8 | ●Runs well 75–112 tok/s | 32K |
| Q5_K_Mmaker’s file | 5.1 GB | ≈ 6.0 | ●Runs well 85–128 tok/s | 32K |
| Q5_0maker’s file | 5.0 GB | ≈ 5.9 | ●Runs well 87–131 tok/s | 32K |
| Q4_K_Mmaker’s file | 4.4 GB | ≈ 5.3 | ●Runs well 98–147 tok/s | 32K |
| Q4_0maker’s file | 4.1 GB | ≈ 5.1 | ●Runs well 103–154 tok/s | 32K |
| INT4 (AWQ / GPTQ)maker’s file | 5.2 GB | ≈ 6.1 | ●Runs well 83–125 tok/s | 32K |
| IQ4_XS | 3.8 GB | ≈ 4.7 | ●Runs well 112–167 tok/s | 32K |
| Q3_K_Mmaker’s filequality drops | 3.5 GB | ≈ 4.5 | ●Runs well 118–177 tok/s | 32K |
| IQ3_XXSquality drops | 2.7 GB | ≈ 3.7 | ●Runs well 149–223 tok/s | 32K |
| Q2_Kmaker’s filequality drops | 2.8 GB | ≈ 3.7 | ●Runs well 145–217 tok/s | 32K |
| IQ2_XXSquality drops | 1.8 GB | ≈ 2.8 | ●Runs well 207–311 tok/s | 32K |
Click a row to choose a precision. “Maker’s file” sizes are the maker’s own files; others are parameters × bits per weight.
Compared with other hardware
Each at the highest precision it runs. Click one for details.
Smallest that runs it (Q4_K_M, 8K): GPU GeForce RTX 3080; unified memory Apple M5 · 16 GB.
Run commands
vllm serve Qwen/Qwen2.5-7B-Instruct --max-model-len 8192
Memory to fine-tune
Batch 1, 2,048-token sequences, gradient checkpointing; LoRA and QLoRA train rank-16 adapters on the attention projections.
Estimates; facts from the maker’s Hugging Face repositories, as of 2026-11-06. How it’s estimated
Cited scores
From third-party leaderboards, not measured by us- Epoch AI · GPQA Diamond 35.5%
#154 of 168 listed models · ±2.29
- Epoch AI · OTIS Mock AIME 2024–2025 2.5%
#149 of 159 listed models · ±0.94
- Epoch AI, 'Capabilities & benchmarking'. Published online at epoch.ai. Retrieved from 'https://epoch.ai/benchmarks'. Licensed CC BY 4.0. · License · as of Nov 4, 2026
- changes we made
Changes
All changes →- Now priced on Alibaba Cloud Model Studio, China Input ¥0.50 · Output ¥1.00
- No longer priced on Alibaba Cloud Model Studio, International Input $0.175 · Output $0.70
- Now priced on Alibaba Cloud Model Studio, International Input $0.175 · Output $0.70
Sources
- Text generation models - Qwen, DeepSeek, GLM | Model Studio - Alibaba Cloud Model Studio - Alibaba Cloud Documentation Center (opens in a new tab) docs.modelstudio.console.alibabacloud.com · checked Oct 15, 2026
- models.dev: qwen2-5-7b-instruct (community data, MIT) (opens in a new tab) github.com · checked Sep 30, 2026
- Alibaba Cloud Model Studio pricing (opens in a new tab) alibabacloud.com · checked Sep 30, 2026
- Qwen2.5: A Party of Foundation Models! | Qwen (opens in a new tab) qwenlm.github.io · checked Sep 30, 2026
- Qwen2.5-7B-Instruct (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen2.5-7B-Instruct model card (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen2.5-7B-Instruct config.json (opens in a new tab) huggingface.co · checked Nov 3, 2026
- Qwen/Qwen2.5-7B-Instruct files (opens in a new tab) huggingface.co · checked Nov 3, 2026
- Qwen/Qwen2.5-7B-Instruct-AWQ files (opens in a new tab) huggingface.co · checked Nov 1, 2026
- Qwen/Qwen2.5-7B-Instruct-GGUF files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen2.5-7B-Instruct-GPTQ-Int4 files (opens in a new tab) huggingface.co · checked Oct 31, 2026
- Qwen/Qwen2.5-7B-Instruct-GPTQ-Int8 files (opens in a new tab) huggingface.co · checked Nov 3, 2026
- How to check model training and deployment pricing-Alibaba Cloud Model Studio(Model Studio)-阿里云帮助中心in Chinese (opens in a new tab) help.aliyun.com · checked Nov 1, 2026
- Qwen2.5-7B-Instruct (opens in a new tab) huggingface.co · checked Oct 31, 2026
Alibaba Qwen’s official policies: data use, privacy, terms →
Some values on this page are still waiting to be checked against official pages: a filled dot is verified, a hollow one is not yet (with or without a source on file). How we verify data
Alibaba Qwen · China · qwen.ai
Related models
Snapshots, the same family and other models from this provider