Qwen3 30B A3B Thinking 2507
Qwen · released Jul 29, 2025 · Qwen/Qwen3-30B-A3B-Thinking-2507
- Type
- Open weightsApache License 2.0
- Params
- 30.5B
- Context
- 82K
3B active per word · about 61K words of context
Our take
Written Aug 2, 2026Qwen3 is a thinking model from Alibaba with 30.5 billion total parameters but only 3 billion active per forward pass, released under an Apache licence. It is designed for step-by-step reasoning at low active-parameter cost, though no benchmark scores are available to verify its quality.
Pick this for self-hosted reasoning workloads where a permissive Apache licence matters, or when you need a 30-billion-parameter-class model with very low active-parameter cost per request. Use it if an 81,920-token context limit fits your task. Skip it if you need measured quality data to make a decision, if your workload is output-heavy given the sharp output price jump, or if you need multimodal input beyond text.
The case for it
- Only 3 billion active parameters from 30.5 billion total — roughly a ten-to-one sparsity ratio that keeps inference cost down.
- Apache 2.0 licence allows commercial use, fine-tuning and redistribution without restriction.
- Cheapest input offer among its own tracked variants.
The case against it
- No benchmark scores in our data: chat, reasoning and coding quality are all unverified.
- Output costs roughly twelve to sixteen times more than input, a steep jump for generation-heavy workloads.
- Throughput is nearly identical across the two measured offers — 147 versus 144 tokens per second, a 2% difference — with the third offer unverified.
How good is it?
We hold no score for this model.
We look for every model we track on every board we watch, and none of them has turned up Qwen3 30B A3B Thinking 2507 — so there is no intelligence, coding, agentic or writing score to show you, not a low one, none.
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
Borderline fit on an estimated size. It leaves 1.6 GB spare on a size we calculated rather than measured, and a 10% error either way would change the answer.
Comfortable fit
GeForce RTX 5090 · 32 GB
Room to spare. 9.6 GB spare means a 10% error in the size would not change the answer.
Apple M1 Pro (16-core GPU) · 32 GB
Room to spare. 2.8 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 3 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
- $0.13 in / $1.56 out
- Context served
- 82K
- Throughput
- ~147 tok/s
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| Alibaba Cloud | $0.13 / $1.56 | 82K | 147 tok/s | No | Yesunknown period | Unknown |
| OpenRouter | $0.20 / $2.40 | 82K | not measured | Unknown | Unknown | Unknown |
| Alibaba Cloudfp8 | $0.20 / $2.40 | 82K | 143 tok/s | No | Yesunknown period | Unknown |
OpenRouterAcross the 3 listings we hold: 2 say they do not train on prompts, 0 say they do and 1 do not say. 0 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 Cloud | ✓ | ✓ | ✗ |
| OpenRouter | ✓ | ✓ | ✗ |
| Alibaba Cloudfp8 | ✓ | ✓ | ✗ |
Tool calling: 3 of 3 listings say yes. JSON output: 3 of 3 listings say yes. Strict schema: 0 of 3 listings say yes, 3 say no.
When we formed this view
Dates behind this page
Prices last checked 3d 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.
- No board we watch has turned up a score, so we hold no quality figures at all.
- Nothing we hold says whether an endpoint streams, so we do not show it either way.
- 1 of 3 listings do not say whether they train on prompts.
- We hold no cached-input rate for any of its listings.
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/Qwen3-30B-A3B-Thinking-2507
- Architecture
- Mixture of experts
- Modality record
- text->text
- Catalogue slug
- qwen-qwen3-30b-a3b-thinking-2507