Qwen3 Next 80B A3B Instruct
Qwen · released Sep 9, 2025 · Qwen/Qwen3-Next-80B-A3B-Instruct
- Type
- Open weightsApache License 2.0
- Params
- 81.3B
- Context
- 262K
3B active per word · about 197K words of context
Our take
Written Aug 3, 2026Qwen3 Next is a downloadable text model from Alibaba with an Apache licence and a 262,144-token request limit. Only 3 billion of its 81.3 billion parameters are active per word, making it a high-capacity, low-activation option that scores best on coding tasks.
Pick this when you need a permissive open licence with a very large request limit, or when coding is the main workload — its Arena Coding score leads its other measured skills by a wide margin. Use it if you want MoE efficiency with only 3 billion active parameters per token. Skip it if creative writing or instruction following is central, or if you need balanced output-to-input pricing.
The case for it
- Strongest measured skill is coding, with an Arena Coding Elo 44.29 points above its own general text score and 129.69 points above its creative writing score.
- Dramatic parameter efficiency: only 3 billion active per token from 81.3 billion total, a 27.1:1 ratio.
- Apache 2.0 licence allows commercial use, fine-tuning and redistribution.
- Wide request limit for the active parameter count: 262,144 tokens with only 3 billion active per forward pass.
The case against it
- Creative writing is its weakest measured category, 85.41 points below its own general text score and 129.69 points below coding.
- Throughput varies widely and is unverified on several providers.
- Output pricing is steep relative to input on every provider, with output costing more than eight times input at every host.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)75th of 143 · 1401.2
CodingWriting and fixing code on its own
Arena Coding75th of 143 · 1445.5
Arena Coding is the only board that has scored it for this.
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Qwen3 Next 80B A3B Instruct for this. We would take the rating from Arena Agent (IPS).
WritingWe do not rate this
Two boards come close and neither tests writing: Arena Creative Writing asks people which of two replies they prefer, and LiveBench Language tests whether a model understood a passage. So we show where Qwen3 Next 80B A3B Instruct placed and give it no mark out of five.
These tests check whether a model follows instructions — a precondition for all the work above, but not a measure of how well that work is done, which is why they get no rating.
Every published score for this model6 scoresEvery figure we hold, from 6 boards, with who ran it and a link to the source — including the boards no rating above is built on.
Can you run it yourself?
- Fits in memory
- weights load entirely on the card
- Spills to system RAM
- some weights offload; much slower
- Too large
- will not load even with offload
- est
- size is calculated; the verdict could change by 10%
GeForce RTX 4090 · 24 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Apple M1 Pro (16-core GPU) · 32 GB
Too large for this card. The weights do not fit even with part of them offloaded to system memory.
Comfortable fit
Apple M2 Max (38-core GPU) · 96 GB
Room to spare. 17.8 GB spare means a 10% error in the size would not change the answer.
Memory use by level
Against a 24 GB card.
All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.
Check against your own machine → · All 71 devices, with every size →
Or rent it from someone else
Cheapest of 9 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.
- per 1M tokens
- $0.10 in / $1.10 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Alibaba Cloudfp8 | $0.098 / $0.78 | 131K | 44 tok/s | No | Yesunknown period | Unknown |
| Alibaba Cloud | $0.098 / $0.78 | 131K | 86 tok/s | No | Yesunknown period | Unknown |
| DeepInfrafp8 | $0.090 / $1.10 | 262K | 51 tok/s | No | No | Confirmed |
| OpenRouter | $0.10 / $1.10 | 262K | not measured | Unknown | Unknown | Unknown |
| DeepInfrafp8 | $0.090 / $1.10 | 262K | not measured | Unknown | Unknown | Unknown |
| Parasailfp8 | $0.10 / $1.10 | 262K | 73 tok/s | No | No | Confirmed |
| Google Vertex AIglobal | $0.15 / $1.20 | 262K | 51 tok/s | No | No | Confirmed |
| Novita AIbf16 | $0.15 / $1.50 | 131K | 28 tok/s | No | No | Confirmed |
| Novita AI | $0.15 / $1.50 | 131K | not measured | Unknown | Unknown | Unknown |
DeepInfra
Google Vertex AI
Novita AIAcross the 9 listings we hold: 6 say they do not train on prompts, 0 say they do and 3 do not say. 4 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.
What each host's API supports
From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.
- ✓
- Supported
- ✗
- Not supported
- Not published
- host gave no parameter list
| Provider | Tool calling | JSON output | Strict schema |
|---|---|---|---|
| Alibaba Cloudfp8 | ✓ | ✓ | ✗ |
| Alibaba Cloud | ✓ | ✓ | ✗ |
| DeepInfrafp8 | ✓ | ✓ | ✓ |
| OpenRouter | ✓ | ✓ | ✓ |
| DeepInfrafp8 | |||
| Parasailfp8 | ✓ | ✓ | ✓ |
| Google Vertex AIglobal | ✓ | ✓ | ✓ |
| Novita AIbf16 | ✓ | ✓ | ✗ |
| Novita AI |
Tool calling: 7 of 9 listings say yes, 2 publish no parameter list. JSON output: 7 of 9 listings say yes, 2 publish no parameter list. Strict schema: 4 of 9 listings say yes, 3 say no, 2 publish no parameter list.
Models people weigh against Qwen3 Next 80B A3B Instruct
When we formed this view
Dates behind this page
Prices last checked 37h ago
What we do not know about this model yet
- We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
- 2 of 9 listings publish no parameter list, so what their API accepts is unknown to us.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 3 of 9 listings do not say whether they train on prompts.
Licence and identifiers
What the licence allowsApache License 2.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.
Licence
Apache License 2.0
Fully permissive: commercial use, redistribution, and derivatives allowed. Requires attribution and a copy of the license. Includes an express patent grant.
Identifiers
- Hugging Face
- Qwen/Qwen3-Next-80B-A3B-Instruct
- Architecture
- Mixture of experts
- Modality record
- text->text
- Catalogue slug
- qwen-qwen3-next-80b-a3b-instruct