// Alibaba (Qwen)
Qwen3-Next-80B-A3B-Instruct
A Alibaba (Qwen) model. 3 prices across 3 channels side by side, each with its source and check date – plus specs, licence, speed and benchmarks.
- Official in/out per 1M
- ¥1 / ¥4
- Lowest (official)
- ¥1 / ¥4
- Context
- 262K
- AA index
- 9.6
Prices checked 2026-10-22Official sources linked on every rowNo estimates
01Specs at a glance
Official values first; gaps filled from OpenRouter’s API where the vendor page is silent.
| Vendor | Alibaba (Qwen) |
|---|---|
| Context window | 262K (262,144 tokens) |
| Max output | 236K |
| Input | Text |
| Output | Text |
| Tool calling / JSON | Yes / Yes |
| Weights & licence | Open weights · Apache-2.0 · Hugging Face |
| Parameters | 81.3B (from weight files) |
| Released | Sep 11, 2025 |
| Model IDs | qwen3-next-80b-a3b-instruct · qwen/qwen3-next-80b-a3b-instruct |
02Prices on 3 channels
Per 1M tokens, in the currency each source publishes (1 USD = 6.7 CNY for the blended column). Sorted: official first, then cheapest.
| Channel | Source | Input | Output | Cache hit | Blended $ | Context | Notes |
|---|---|---|---|---|---|---|---|
| Official Vendor’s China site, billed in CNY | Official 10-10 | ¥1 | ¥4 | — | 0.261 | 262K | Batch −50%Beijing region; batch -50% on supported models; explicit cache create 125% / hit 10% of input |
| DeepInfra Inference platform (own price page) | Platform 10-10 | $0.09 | $1.1 | — | 0.343 | 262K | DeepInfra own pricing page |
| Parasail · fp8 Via OpenRouter / provider listed there | OpenRouter 10-22 | $0.1 | $1.1 | $0.07 | 0.35 | 262K | Parasail price as listed on OpenRouter (pass-through; quantization=fp8; discount field=0); not checked against Parasail's own page |
“Official” = vendor pricing page; “Platform” = the platform’s own pricing page; “OpenRouter” = price listed on OpenRouter, not checked on the platform’s site. Click a source tag to open the original page.
03Where you can use it
Derived only from the price sources above. Country restrictions are set by each vendor – check their terms before you sign up.
- Vendor’s China site, billed in CNY
Official - Inference platform (own price page)
DeepInfra - Via OpenRouter / provider listed there
Parasail - Self-host the open weights
GPU / VRAM sizing and build plans→
04Performance
Speed and AA index from Artificial Analysis public leaderboards (not our own tests); benchmark scores from our benchmark library, best result per benchmark.
- AA indexQwen3 Next 80B A3B Instruct
- 9.6
- Output speed (median)
- 115 tok/s
- Time to first token
- 0.9s
- AA index per $ (blended)
- 36.92
| Benchmark | Best score | Source | Date |
|---|---|---|---|
| LMArena Text | 1399 | official | 2026-10-08 |
05Price changes we recorded
All price changes →From our daily re-checks of the sources above.
No price change recorded since we started tracking (2026-10-10).
06More from Alibaba (Qwen)
FAQ
How much does the Qwen3-Next-80B-A3B-Instruct API cost?
On the official pricing page: ¥1 / ¥4 per 1M input / output tokens (checked 2026-10-10).
What is the context window of Qwen3-Next-80B-A3B-Instruct?
262K tokens of context, up to 236K output tokens. Platforms may cap it lower – see the table.
Is Qwen3-Next-80B-A3B-Instruct open source?
Yes, the weights are published on Hugging Face (Qwen/Qwen3-Next-80B-A3B-Instruct). Licence: Apache-2.0.