Qwen3.5-27B
Qwen · released Feb 24, 2026 · Qwen/Qwen3.5-27B
- Type
- Open weightsApache License 2.0
- Params
- 27.8B
- Context
- 262K
about 197K words of context
Our take
Written Aug 2, 2026Qwen3.5-27B is a 27.8-billion-parameter multimodal model from Alibaba that accepts text, images and video under a permissive Apache licence. It handles up to 262,144 tokens in a single request and shows its strongest measured results on coding tasks.
Pick this for coding-heavy workloads where its leaderboard score peaks, or for long-context applications needing a quarter-million-token window. Use it if you want open weights with a genuinely permissive licence for commercial use, fine-tuning or redistribution. Skip it if creative writing quality matters — that is its weakest measured category — or if you need web-specific coding, which trails its general coding score by a wide margin.
The case for it
- Strong coding performance relative to general chat: its coding leaderboard score exceeds its overall text score by more than 41 points.
- Permissive Apache 2.0 licence with full multimodal inputs including video, and 27.8 billion parameters suitable for local deployment.
- Extremely long context window for its parameter class: 262,144 tokens.
- Broad benchmark coverage across seven distinct Arena categories, with maths and hard-prompt scores near its coding peak.
The case against it
- Creative writing is a clear relative weakness: more than 92 points below its coding score and over 50 points below its overall text score.
- Web-specific coding lags general coding by about 92 points on the leaderboard.
- No disclosed active parameter count: the full 27.8 billion parameters must load and run per token.
How good is it?
IntelligencePuzzles, maths, exam questions
Arena Text (overall)72nd of 143 · 1408
CodingWriting and fixing code on its own
Arena Coding74th of 143 · 1449.6
AgenticPlanning, calling tools, staying on task
Nobody we watch has scored Qwen3.5-27B 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 Qwen3.5-27B 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 model7 scoresEvery figure we hold, from 7 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%
Comfortable fit
GeForce RTX 4090 · 24 GB
Room to spare. 3 GB spare means a 10% error in the size would not change the answer.
GeForce RTX 5090 · 32 GB
Room to spare. 11 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. 4.2 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 10 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
- $0.20 in / $1.56 out
- Context served
- 262K
- Throughput
- Not measured
| Provider | In / out per 1M tokens | Context | Throughput | Trains on prompts | Logs prompts | Zero retention |
|---|---|---|---|---|---|---|
| OpenRouter | $0.20 / $1.56 | 262K | not measured | Unknown | Unknown | Unknown |
| Alibaba Cloudfp8 | $0.20 / $1.56 | 262K | 21 tok/s | No | Yesunknown period | Unknown |
| Alibaba Cloud | $0.20 / $1.56 | 262K | 23 tok/s | No | Yesunknown period | Unknown |
| SiliconFlowfp8 | $0.25 / $2.00 | 262K | 10 tok/s | No | No | Confirmed |
| AtlasCloudfp8 | $0.27 / $2.16 | 262K | 15 tok/s | No | Yesunknown period | Unknown |
| Novita AIbf16 | $0.30 / $2.40 | 262K | 13 tok/s | No | No | Confirmed |
| Novita AI | $0.30 / $2.40 | 262K | not measured | Unknown | Unknown | Unknown |
| Phala | $0.30 / $2.40 | 262K | 54 tok/s | No | No | Confirmed |
| DeepInfrafp8 | $0.26 / $2.60 | 262K | 49 tok/s | No | No | Confirmed |
| DeepInfrafp8 | $0.26 / $2.60 | 262K | not measured | Unknown | Unknown | Unknown |
Novita AI
DeepInfraAcross the 10 listings we hold: 7 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 |
|---|---|---|---|
| OpenRouter | ✓ | ✓ | ✓ |
| Alibaba Cloudfp8 | ✓ | ✓ | ✓ |
| Alibaba Cloud | ✓ | ✓ | ✓ |
| SiliconFlowfp8 | ✓ | ✓ | ✓ |
| AtlasCloudfp8 | ✓ | ✓ | ✓ |
| Novita AIbf16 | ✓ | ✓ | ✗ |
| Novita AI | |||
| Phala | ✗ | ✓ | ✓ |
| DeepInfrafp8 | ✓ | ✓ | ✓ |
| DeepInfrafp8 |
Tool calling: 7 of 10 listings say yes, 1 says no, 2 publish no parameter list. JSON output: 8 of 10 listings say yes, 2 publish no parameter list. Strict schema: 7 of 10 listings say yes, 1 says no, 2 publish no parameter list.
Models people weigh against Qwen3.5-27B
When we formed this view
Dates behind this page
Prices last checked 38h 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 10 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 10 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/Qwen3.5-27B
- Architecture
- Dense
- Modality record
- text+image+video->text
- Catalogue slug
- qwen-qwen3-5-27b