Multimodal AI has grown from novelty to a should in latest occasions. Want proof? If I have been to inform you to work on an AI mannequin that solely understands textual content, you’ll most likely snort and throw 10 mannequin names at me that may work throughout codecs – be it textual content, audio, or visuals. The brand new race, thus, for the bigwigs, is to not make simply one other AI mannequin, however a system that may perceive the world extra like people do. That is naturally finished by language, visuals, sound, and movement collectively. That’s the house Alibaba’s new Qwen3.5-Omni enters.
The newest mannequin in Alibaba’s Qwen household is positioned as a “absolutely omni-modal LLM”. We will discover what meaning in idea, and what this moniker guarantees within the practicality of issues. One factor is for positive (with Qwen and different launches just like the latest Gemini 3.1 Flash Dwell), AI fashions have gotten much less like separate instruments and extra like unified interactive methods.
For now, we concentrate on the Qwen3.5-Omni and all that it brings ot the desk.
What’s Qwen3.5-Omni?
As I discussed earlier, this one is a completely omni-modal mannequin underneath the Qwen household. In easy phrases, it’s constructed to deal with textual content, photographs, audio, and audio-visual content material inside a single system. That’s what separates it from older AI setups, the place every modality usually wanted a unique mannequin or pipeline.
As is clear from its launch transient, Alibaba is pitching Qwen3.5-Omni as a mannequin designed for richer, extra pure interplay with real-world inputs. As a substitute of treating voice, photographs, and video as non-compulsory add-ons, it presents them as core elements of the mannequin itself. Meaning Qwen3.5-Omni is far more than a normal chatbot. It’s a multimodal AI system meant to interpret completely different sorts of data collectively.
As for its variants, the brand new Qwen3.5-Omni collection contains Instruct variants in three sizes – Plus, Flash, and Mild. This household construction makes it ultimate for various use circumstances and efficiency wants. The launch additionally highlights long-context assist, which suggests the mannequin will not be solely broad in modality but in addition constructed for heavier, extra sustained inputs.
There are, after all, extra such options in line. Right here is all that the brand new Qwen3.5-Omni brings to the desk.
Qwen3.5-Omni Options
Qwen3.5-Omni is clearly a extra succesful step up from Qwen3-Omni. Although the factor to notice right here is that it comes with a lot broader horizons as effectively. Right here is how:
1. Stronger multilingual capabilities
In contrast with Qwen3-Omni, Qwen3.5-Omni comes with considerably improved multilingual capabilities, together with speech recognition in 113 languages.
2. Lengthy-context assist
The Qwen3.5-Omni collection contains Instruct variations with assist for 256K long-context enter. This factors to a mannequin that’s designed for a lot bigger and extra sustained prompts than a normal chatbot workflow.
3. A number of mannequin sizes
The collection contains three Instruct sizes: Plus, Flash, and Mild. That offers Qwen3.5-Omni a extra versatile product household fairly than a single one-size-fits-all launch.
4. Giant multimodal enter capability
The announcement weblog says the mannequin can deal with greater than 10 hours of audio enter and over 400 seconds of 720p audio-visual enter at 1 FPS. That signifies that it’s constructed for heavier audio and video understanding workloads.
5. Semantic interruption assist
Qwen3.5-Omni helps semantic interruption by “native turn-taking intent recognition.” In easy phrases, this helps the mannequin distinguish between significant consumer interruption and irrelevant background noise. All in all, the characteristic makes dwell conversations really feel extra pure.
6. Native WebSearch and Perform Calling
The mannequin natively helps WebSearch and complicated FunctionCall capabilities. This permits it to determine by itself whether or not it ought to invoke WebSearch with the intention to reply a consumer’s real-time query. Assume extra agent-like sensible use.
7. Finish-to-end voice management and dialogue
This can be a very attention-grabbing improve with Qwen3.5-Omni, and I’m positive you’ll like it too when you see its demos. The brand new Qwen mannequin helps end-to-end voice management and dialogue. This implies the mannequin can comply with spoken directions in a extra human-like approach by controlling elements of speech reminiscent of quantity, velocity, and emotion. As demonstrated in some movies, the mannequin can whisper, shout, and even categorical feelings in a approach that can sound very pure to most.
8. Voice cloning
One other notable characteristic is voice cloning, which permits customers to add a voice and customise the AI assistant’s output voice accordingly. It means now you can communicate to the AI and have it reply within the voice of your selection.
With all these options, right here is how the Qwen3.5-Omni performs in benchmark exams.
Qwen3.5-Omni: Benchmark Efficiency
Slightly than profitable each single benchmark outright, Qwen3.5-Omni-Plus comes throughout as a really well-rounded omni-modal mannequin that stays extremely aggressive throughout textual content, imaginative and prescient, audio, audio-visual understanding, and speech era. That’s the larger takeaway right here: consistency throughout virtually each format. And as an add-on, it both leads or comes extraordinarily shut usually occasions to the highest mannequin within the comparability.
1. Audio: USP of the mannequin
Audio is clearly one in all Qwen3.5-Omni-Plus’s strongest areas.
In audio understanding, it barely edges out Gemini-3.1-Professional on MMAU (82.2 vs 81.1) and MMSU (82.8 vs 81.3), whereas additionally delivering an enormous leap on RUL-MuchoMusic (72.4 vs 59.6). On dialogue-heavy duties, it posts the very best rating on VoiceBench (93.1), forward of Gemini-3.1-Professional’s 88.9.
Its speech-related efficiency can be spectacular in transcription and recognition-style duties. For instance, on LibriSpeech, Qwen3.5-Omni-Plus scores 1.11 / 2.23, forward of Gemini-3.1-Professional’s 3.36 / 4.41, and on CV15 (en) it information 4.83 towards Gemini’s 8.73. That means Qwen is especially robust not simply at listening to audio, however at processing it precisely.
2. Audio-Visible: Robust, however not all the time the outright chief
On audio-visual duties, Qwen3.5-Omni-Plus performs strongly, although that is one space the place Gemini-3.1-Professional nonetheless holds some benefits.
For example, Qwen leads on DailyOmni (84.6 vs 82.7) and QualcommInteractive (68.5 vs 66.2), and likewise tops Omni-Cloze (64.8 vs 57.2) in captioning. However Gemini stays forward on benchmarks like WorldSense (65.5 vs 62.8), VideoMME with audio (89.0 vs 83.7), and OmniGAIA software use (68.9 vs 57.2).
3. Visible: Aggressive, with some category-leading scores
In visible duties, Qwen3.5-Omni-Plus once more appears balanced and succesful fairly than wildly dominant.
It posts the very best rating on MMMU-Professional (73.9), edges forward on RealWorldQA (84.1), leads on CC-OCR (83.4), tops EmbSpatialBench (85.4), and performs greatest on a number of video benchmarks, together with VideoMME with out subtitles (81.9), MLVU (86.8), MVBench (79.0), LVBench (71.2), and MME-VideoOCR (77.0).
That stated, the non-thinking Qwen3.5-Plus baseline nonetheless beats it on some traditional visible and reasoning-heavy benchmarks reminiscent of MMMU, MathVision, and Mathvista mini. So Qwen3.5-Omni-Plus might not be the very best visible mannequin in isolation. Although it nonetheless demonstrates very stable visible efficiency whereas bringing audio and speech into the identical system.
4. Textual content: Stable, however not the headline story
Qwen3.5-Omni-Plus reveals textual content efficiency, although it doesn’t seem like the central headline of the discharge.
Qwen3.5-Omni-Plus stays near the non-thinking Qwen3.5-Plus mannequin on a number of benchmarks: MMLU-Redux (94.2 vs 94.3), C-Eval (92.0 vs 92.3), and IFEval (89.7 vs 89.7). It additionally does fairly effectively on long-context duties like LongBench v2 (59.6) and reasoning duties like HMMT Nov 25 (84.4).
The broader sample is that Qwen3.5-Omni-Plus preserves a robust textual content basis whereas extending into different modalities. In fact, it isn’t probably the most thrilling a part of the benchmark desk. However it’s reassuring that the multimodal enlargement does lower down on text-quality.
5. Speech Technology: Standout benchmark outcomes
This is among the clearest strengths of the mannequin.
In customized voice stability, decrease is healthier, and Qwen3.5-Omni-Plus performs extraordinarily effectively. It scores 1.07 on Seed-zh, beating ElevenLabs (13.08), Gemini-2.5 Professional (2.42), GPT-Audio (1.11), and Minimax (1.19). It additionally leads on Seed-hard (6.24) and performs greatest on the multilingual averages proven, together with 2.06 on Public-Multilingual-avg (20 languages) and 5.82 on Inhouse-Multilingual-avg (9 languages).
On voice clone stability, it additionally posts the very best multilingual rating within the public setting at 1.87, forward of ElevenLabs (10.29) and Minimax (2.52). On voice clone similarity, larger is healthier, and Qwen3.5-Omni-Plus reaches 0.79 and 0.80, which is once more the strongest rating within the comparability proven.
This makes speech era probably the most compelling elements of the Qwen3.5-Omni-Plus benchmark story.
Total takeaway
- Strongest: Audio, Speech era
- Very robust: Audio-visual, Imaginative and prescient
- Stable/above common: Textual content
This efficiency is made doable due to the distinctive structure of the Qwen mannequin. Right here is
Qwen3.5-Omni Structure
Qwen3.5-Omni follows what Qwen calls a Thinker-Talker structure. Now we have seen it earlier than within the earlier Qwen fashions. As a substitute of treating understanding and response era as one blended course of, the mannequin separates them into two practical elements. That makes the structure simpler to know, particularly for a mannequin constructed to deal with a number of modalities.
Here’s what each elements do:
1. The Thinker
The Thinker is answerable for the mannequin’s understanding layer. In accordance with Qwen, it receives visible and audio indicators by the mannequin’s encoders and handles the higher-level reasoning over these inputs.
In easy phrases, that is the a part of the system that interprets what the mannequin is seeing, listening to, or studying earlier than a response is generated.
2. The Talker
The Talker handles the output facet of the system. As soon as the mannequin has processed the enter, this part is answerable for producing the response.
This distinction issues as a result of Qwen3.5-Omni isn’t just meant to analyse inputs. Additionally it is meant to reply to interactive and conversational use circumstances.
3. Hybrid-Consideration MoE in Each Elements
Qwen says that each the Thinker and the Talker undertake Hybrid-Consideration MoE.
That element suggests the structure is designed to steadiness functionality and effectivity. As a substitute of counting on one giant block to handle all the things, the mannequin makes use of a extra structured design to assist each multimodal understanding and response era.
Why This Structure Issues
For an omni-modal mannequin, structure issues greater than traditional. Qwen3.5-Omni is anticipated to course of textual content, photographs, audio, and audio-visual content material inside one system. A cut up between understanding and era helps assist that broader position.
That is additionally why, fairly than trying like a textual content mannequin with a number of added multimodal options, Qwen3.5-Omni is being framed as a system designed from the bottom up for richer interplay throughout completely different enter and output modes.
Now that we all know the way it works, right here is the right way to entry the brand new Qwen mannequin.
Qwen3.5-Omni: Learn how to Entry
There are 3 important methods to entry the Qwen3.5-Omni, principally primarily based in your use case. These are:
1. Qwen Chat
Probably the most simple approach to attempt Qwen3.5-Omni is thru Qwen Chat, which acts because the direct user-facing entry level for the mannequin household.
Finest for: particular person customers
2. through Offline API in Alibaba Cloud Mannequin Studio
For traditional API-based integration, Alibaba Cloud supplies Qwen-Omni by Mannequin Studio. The mannequin accepts textual content mixed with one different modality right here, reminiscent of picture, audio, or video, and may generate responses in textual content or speech. Alibaba notes that Qwen-Omni at present helps OpenAI-compatible calls solely, requires an API key, and works with the most recent SDK.
Finest for: app integration and multimodal era workflows
3. through Realtime API for dwell audio and video interactions
For interactive purposes, Alibaba Cloud additionally gives Qwen-Omni-Realtime, which is accessed by a stateful WebSocket connection. This route is supposed for real-time audio and video chat use circumstances, the place the mannequin can course of streaming inputs and generate responses repeatedly throughout a session.
Finest for: voice- or video-driven dwell experiences
Qwen3.5-Omni: Demonstration
The Qwen group has shared a number of demos of the brand new Qwen3.5-Omni that showcase its capabilities throughout use circumstances. Test them out beneath:
1. Audio-Visible Captioning
The primary demo for the mannequin is that of audio-visual captioning. The demo reveals how the mannequin is ready to precisely interpret the knowledge being shared inside a video and generate the textual content for a similar. Test it out in motion within the embed beneath.
2. Audio-Visible Vibe Coding
This one is tremendous attention-grabbing, because it reveals Qwen3.5-Omni decoding particular technical directions shared inside a video, after which performing accordingly. As might be seen, the mannequin can clearly perceive what is occurring throughout visible and audio inputs and help in producing or refining code accordingly. That is combining multimodal context into the coding loop, making the interplay really feel extra intuitive than a plain text-only workflow.
3. Multi-Flip Dialogue and Clever Interruption
Alibaba additionally shares proof for its claims of multi-turn dialogue capabilities on the Qwen3.5-Omni. In one other video, the mannequin might be seen dealing with interruptions tremendous intelligently. It showcases that the Qwen3.5-Omni can casually maintain a back-and-forth dialog whereas additionally recognising when a consumer is meaningfully interrupting, as an alternative of reacting awkwardly to each sound or pause.
The anchor might be clearly seen attempting to idiot the mannequin with filler phrases like “hmmm” and “okay” in the course of the mannequin’s response. Although Qwen3.5-Omni appears to know higher than to interrupt.
4. Voice Model, Emotion, and Quantity Management
Should you have been to ask me, this appears to be the USP of the brand new Qwen mannequin. Now we have all seen AI fashions conversate with us in a really related (if not precise) tone as people. The Qwen3.5-Omni now takes it a step additional and brings in voice type, emotion, and quantity management. The demo highlights how the mannequin can whisper, shout, and even narrate a poem whereas feeling dejected. That’s one thing you don’t see too usually.
Conclusion
From what we will see within the demos and the knowledge shared by Alibaba, the brand new Qwen3.5-Omni takes the multi-modal capabilities of an LLM to a different degree. From deciphering audio-visual directions to creating AI conversations really feel far more human, it brings with it a set of options which are not often seen in AI fashions.
I’m positive many would love to change to Qwen3.5-Omni after this, largely for the whole conversations occurring in audio-visual inputs and outputs. Whether or not they ship on the standard that’s showcased right here, stays to be seen.
Login to proceed studying and revel in expert-curated content material.