Qwen3 14B
Qwen3 is the latest generation of large language models in Qwen series, offering a comprehensive suite of dense and mixture-of-experts (MoE) models.
text → text · made by Alibaba
Sold by 12 ways
| Seller | Lane | Rate |
|---|---|---|
| Hugging Face Inference Providers 2 ways | $0.07per Mtok in$0.2per Mtok out | |
| Hugging Face Inference Providers aggregator | nscale | $0.07per Mtok in$0.2per Mtok outrouter.huggingface.co · read 2026-08-25 |
| Hugging Face Inference Providers aggregator | deepinfra | $0.12per Mtok in$0.24per Mtok outrouter.huggingface.co · read 2026-08-25 |
| SiliconFlow aggregator | standard | $0.07per Mtok in$0.28per Mtok outmodels.dev · read 2026-09-16 |
| Nous Research aggregator | standard | $0.1per Mtok in$0.22per Mtok outinference-api.nousresearch.com · read 2026-09-12 |
| OpenRouter 4 ways | $0.12per Mtok in$0.24per Mtok out | |
| OpenRouter aggregator | nextbit/int4 | $0.1per Mtok in$0.22per Mtok outopenrouter.ai · read 2026-08-25 |
| OpenRouter aggregator | standard | $0.12per Mtok in$0.24per Mtok outopenrouter.ai · read 2026-09-14 |
| OpenRouter aggregator | deepinfra/fp8 | $0.12per Mtok in$0.24per Mtok outopenrouter.ai · read 2026-09-03 |
| OpenRouter aggregator | alibaba | $0.23per Mtok in$0.91per Mtok outopenrouter.ai · read 2026-08-25 |
| DeepInfra aggregator | standard | $0.12per Mtok in$0.24per Mtok outapi.deepinfra.com · read 2026-08-25 |
| Vercel AI Gateway aggregator | standard | $0.12per Mtok in$0.24per Mtok outai-gateway.vercel.sh · read 2026-08-25 |
| Fireworks AI aggregator | standard | $0.2per Mtok in$0.2per Mtok outraw.githubusercontent.com · read 2026-09-16 |
| Inference.net aggregator | standard | $0.23per Mtok in$0.91per Mtok outapi.inference.net · read 2026-09-16 |
Measured 11 standings
| Place | Board | Metric | Score |
|---|---|---|---|
| 17thof 105 | Vectara Hallucination Leaderboard | Hallucination Rate | 5.4 |
| 91stof 123 | Lmca — Epoch AI | Score | 18.19 |
| 101stof 161 | Dtbench — Epoch AI | Accuracy | 64 |
| 132ndof 143 | Artificial Analysis · Agentic Index | Index | 0.9 |
| 146thof 290 | OTIS Mock AIME 2024-2025 | mean_score | 66.389 |
| 149thof 168 | Scicode — Epoch AI | Score | 31.597 |
| 165thof 222 | Chess puzzles — Epoch AI | mean_score | 4 |
| 166thof 184 | Artificial Analysis · Coding Index | Index | 13.8 |
| 176thof 177 | Critpt — Epoch AI | Accuracy | 0 |
| 198thof 313 | GPQA Diamond | mean_score | 63.763 |
| 469thof 656 | Epoch capabilities index — Epoch AI | ECI Score (none) | 138.26 |
About
Qwen3 14B — a text model from Alibaba, sold by 8 companies from $0.07 in and $0.28 out per million tokens, placed 17th of 105 on Vectara Hallucination Leaderboard.
It takes text and returns text, with a context window of 40,960 tokens. It was published in April 2025, trained on material up to April 2025. Its sellers say it can reason step by step and call a tool. The catalogue files it under reasoning. Eight companies sell it. The cheapest is $0.07 in and $0.28 out per million tokens at SiliconFlow. Beside the standard rate there are separately routed and regional lanes. It has been measured on 11 boards, and stands best at 17th of 105 on Vectara Hallucination Leaderboard.
Every current figure
- Maker
- Alibaba
- Register
- model
- Takes
- text
- Returns
- text
- Context
- 40,960 tokens
- Published
- April 2025
- Knowledge to
- April 2025
- Parameters
- 15 billion
- Licence
- not read
- Sellers
- 8
- Maker's own price
- not read
- Price
- $0.07 in and $0.28 out per million tokens — SiliconFlow
- Boards
- 11
- Best place
- 17th of 105 — Vectara Hallucination Leaderboard
Known as 2 names
Qwen/Qwen3-14Balibaba/qwen-3-14b