QwQ Base: pricing, context window and benchmarks
QwQ Base is a multimodal model from Alibaba Cloud. It offers a 65,536-token context window, costs $3.68 per million input tokens and $15.98 per million output tokens, and averages 74.1 across the index's benchmark suite.
Pricing
QwQ Base costs $3.68 per million input tokens and $15.98 per million output tokens. A blended 3:1 input-to-output mix works out to about $6.76 per million tokens. Cached input is listed at $0.32 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $3.68 | per 1M tokens |
| Output | $15.98 | per 1M tokens |
| Cached input | $0.32 | per 1M tokens |
| Blended 3:1 | $6.76 | estimate per 1M tokens |
| 1M in + 1M out | $19.66 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 77.5 | 263 | 56th | |
| GPQA Diamond | 69.9 | 268 | 55th | |
| SWE-bench Verified | 65.2 | 211 | 65th | |
| HumanEval | 77.0 | 308 | 49th | |
| MATH-500 | 77.5 | 266 | 56th | |
| AIME 2025 | 66.2 | 326 | 46th | |
| LiveCodeBench | 64.0 | 319 | 47th | |
| IFEval | 81.2 | 303 | 50th | |
| TruthfulQA | 79.8 | 190 | 68th | |
| HellaSwag | 86.4 | 201 | 67th | |
| MMMU | 70.0 | 304 | 49th | |
| Arena Elo | 1373.8 | 300 | 50th |
What QwQ Base is good for
QwQ Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.68 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.
Strengths
- Input price of $3.68 per million tokens is at or below the index median.
- Open weights make self-hosting, fine-tuning and data-residency review possible.
- Accepts more than text, including image input where the provider supports it.
Trade-offs
- No standout weakness in the indexed specifications, but real-world behaviour still needs testing.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| DeepSeek-Coder Base | DeepSeek | 32768 | $3.53 | 74.1 |
| GPT-5.1 Nano | OpenAI | 32768 | $9.56 | 74.1 |
| Gemma 3 Instruct | Google DeepMind | 128000 | $3.72 | 74.1 |
| Nova Lite Pro | Amazon | 200000 | $9.69 | 74.1 |
| QwQ Base | Alibaba Cloud | 65536 | $3.68 | 74.1 |
| Reka Flash | Reka AI | 128000 | $9.95 | 74.1 |
FAQ
How much does QwQ Base cost?
QwQ Base is listed at $3.68 per million input tokens and $15.98 per million output tokens. Cached input is $0.32 per million tokens. A blended 3:1 input-to-output mix is about $6.76 per million tokens.
What is the context window of QwQ Base?
QwQ Base has a context window of 65,536 tokens and can emit up to 32,768 output tokens per request in this index.
Is QwQ Base open weights?
Yes. QwQ Base is listed as an open-weight model under a open weights licence.
How does QwQ Base score on benchmarks?
QwQ Base averages 74.1 across the scored benchmarks in this index, ranking #276 of 600 models. Its strongest relative result is Arena Elo at 1373.8.
What is QwQ Base best for?
QwQ Base is an open-weight model, so it can be self-hosted or run through a third-party inference provider. That matters for data residency, fine-tuning and long-run cost control. At $3.68 per million input tokens (hosted list price) and a 65,536-token context window, it is a practical option for teams that want more control than a closed API allows.