Qwen3 Next 80B A3B
AlibabaReleased Dec 1, 2024Open weights · Apache 2.0qwen3-next-80b-a3b
- Reasoning
- no
- Knowledge cutoff
- September 2025
- Context
- 262K tokens
- Max output
- 32k tokens
- Input
- text
- API features
- tools, structured outputs, JSON mode, seed, logprobs
- Alibaba's API
- yes, as qwen3-next-80b-a3b-instruct
Read Oct 23, 06:00 UTC from the hosts' published offers and OpenRouter, spec sheet from Alibaba · cutoff: aiknowledgecutoff.com
The cheapest host, Alibaba, serves a weights precision we can't read (host terms) and serves a 128K context (maker: 256K). Weights and context per host
Run it yourself
open weights, Apache 2.0 · fits 4 of the 6 workstations in our guideSelf-host guide ›- Parameters
- 80B
- 3B active per token (MoE)
- Memory needed
- 48 GB
- in INT4 · 96 GB in INT8
- Smallest cloud setup
- A100
- 80 GB · $1.19/h, about $869 a month
- Worth it above
- 989M tokens
- a month, against the cheapest API at $0.88 per M
| Quantization | Memory | Smallest cloud GPU | Quality |
|---|---|---|---|
| INT4 (Q4) | 48 GB | A100 · $1.19/h | smallest footprint, minor quality loss |
| INT8 (Q8) | 96 GB | 2× H100 · $4.30/h | near-lossless |
| FP16 (full) | 192 GB | 4× H100 · $8.60/h | reference quality |
At home: in INT4 it fits a Mac Studio M5 Ultra (96 GB unified), about $5,499 once, or $0.2092/h over 3 years, electricity aside; it pays for itself after about 6B tokens against the cheapest API.
Memory ≈ parameters × bytes per parameter × 1.2 · cloud price: on-demand rate, running 24/7 · builds: GGUF, AWQ, MLX, GPTQ, Ollama · break-even calculator
| Provider | Input | Output | Cache ↑ | tok/s | First token | Uptime 24h | Checked |
|---|---|---|---|---|---|---|---|
| Parasailfp8 (bf16) · 256K ctx | $0.10 | $1.10 | $0.07 | 139 | 0.46 s | 100% | Oct 23 |
| Alibabaprecision not read: host terms · 128K ctx (256K) · 32K out | $0.0975 | $0.78 | — | — | — | 100% | Oct 9 |
| DeepInfrafp8 (bf16) · 256K ctx · 16K out | $0.09 | $1.10 | — | 110 | 0.55 s | 99.9% | Oct 23 |
| Googleprecision not stated · 256K ctx | $0.15 | $1.20 | — | — | — | 96.3% | Oct 22 |
| Novita AIprecision not stated · 128K ctx (256K) · 32K out | $0.15 | $1.50 | — | — | — | — | Oct 8 |
5 of 5 providers · prices per million tokens, read Oct 23, 06:00 UTC · speed: OpenRouter, median of the last 30 min, read Oct 23, 06:00 UTC · uptime: OpenRouter, last 24 h · under each host: weights precision, context and max output as served, the maker's value in brackets when the host's differs
| Provider | Weights | Context | Max output | Tools | Structured outputs | Reasoning |
|---|---|---|---|---|---|---|
| Alibaba | not read: host terms | 128K | 32K | yes | no | no |
| DeepInfra | fp8 | 256K | 16K | yes | yes | no |
| Parasail | fp8 | 256K | not read: host terms | yes | yes | no |
| not stated by the host | 256K | not stated by the host | yes | yes | no | |
| Novita AI | not stated by the host | 128K | 32K | — | — | — |
Weights, context and max output: read at each host's own catalogue or docs where its terms allow, otherwise as the host declares to OpenRouter; hover a cell for its source and date · tools, structured outputs, reasoning: as declared to OpenRouter · not verified by us
Qwen3 Next 80B A3B spec sheet, field by field, with sources
Qwen3 Next 80B A3B spec sheetmaker-declared
What Alibaba declares about the model itself — read from its own weights repository and documentation, not from any host. What each endpoint actually serves is in the pricing above: the two can differ, and both are true.
- Context window
- 262,144 tokens
- Parameters
- 81,324,862,720≈ 81.3B
- Active experts
- 10 of 512 active per token
- Weight precision
- bfloat16
- License
- Apache 2.0 — open weights, permissivecurated · read 2026-09-07Scope: the licence on the published weights, and nothing else. It is not the maker's acceptable-use policy, which is a separate document, and it is not the contract you sign with whichever host you call — hosts set their own terms, which this index does not read.
- Knowledge cutoff
- not published by the maker≈ September 2025 third-party estimate — not the maker's figurecurated
- Modalities
- text
- Max output (official)
- to verify
- Release date
- to verify≈ December 1, 2024 third-party catalogue — not the maker's figure
- Maker lifecycle status
- to verify
Also declared in the weights
- Architecture class
- Qwen3NextForCausalLM
- Layers
- 48
- Vocabulary
- 151,936 tokens
- Weights first published
- September 9, 2025
- Weights last updated
- September 17, 2025
Qwen/Qwen3-Next-80B-A3B-Instruct · config.json @ 9c7f2fb · read 2026-10-08 · 6 of 10 fields published by the maker
| Date | Provider | Input | Output | Change |
|---|---|---|---|---|
| Oct 22 | DeepInfra | $0.072 → $0.09 | $0.88 → $1.10 | +25% |
| Oct 16 | DeepInfra | $0.09 → $0.072 | $1.10 → $0.88 | −20% |
Every published change of input or output price, newest first · change on input + output · full series: pricing-history.csv.
| Provider | Chatbot 1,000 in / 500 out | RAG / long context 8,000 in / 500 out | Batch summarize 4,000 in / 1,000 out |
|---|---|---|---|
| Alibaba | $0.49 | $1.17 | $1.17 |
| DeepInfra | $0.64 | $1.27 | $1.46 |
| Parasail | $0.65 | $1.35 | $1.50 |
| $0.75 | $1.80 | $1.80 | |
| Novita AI | $0.90 | $1.95 | $2.10 |
What each host bills for 1,000 requests of a representative job
Your own workload
Other Alibaba models · all its models, hosts and history ›
- Qwen3.8 27B16 providers · from $1.29
- Qwen3.6 35B A3B12 providers · from $1.00
- Qwen3 235B A22B Instruct 250710 providers · from $0.44
- Qwen3.5 397B A17B10 providers · from $3.45
- Qwen3.6 27B9 providers · from $2.25
- Qwen3.5-35B-A3B8 providers · from $1.14