Catalogue
Models
455 models, and deliberately not every model there is: one earns a place by being sold by a provider we track, measured by a leaderboard we follow, or published as open weights by a lab we follow. Pointer names like “-latest” stay out, because they answer for whatever the vendor aimed them at that week.
How we choose what to list49 models
Qwen3.8 Omni FlashQwenQwen3.8 Omni Flash is a hosted-only model that takes text, images, audio and video in the same request and returns text.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.15 / $0.47per 1Mvia Alibaba Cloud
Qwen3.83 sizes here
Qwen3.8 FlashQwenQwen3.8 Flash is a hosted-only model with a lopsided profile: it follows instructions very well and builds web apps well, but it is a weak agent.writingnot measuredcodingnot measuredagents29th of 55 everydaynot measured$0.11 / $0.38per 1Mvia DeepInfraQwen3.8 27BQwenQwen3.8 27B is a downloadable model you can run yourself, and its measured strength is maths and reasoning.writing93rd of 168 coding45th of 168 agents33rd of 55 everyday63rd of 168 $0.15 / $1.88per 1Mvia PhalaQwen3.8 MaxQwenQwen3.8 Max is a hosted-only model released in August 2026, with no download listed and no licence terms disclosed.writing12th of 168 coding26th of 168 agents18th of 55 everyday17th of 168 $2.00 / $6.00per 1Mvia Novita AIQwen3.8 2.4T A95BQwenQwen3.8 2.4T A95B is a text-in, text-out model you can download and run yourself, and ten hosts serve it at one shared rate.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$2.00 / $6.00per 1Mvia OpenRouterQwen3.73 sizes here
Qwen3.7 FlashQwenQwen 3.7 Flash is a proprietary multimodal model from Alibaba that handles text, images and video with a one-million-token request limit.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.030 / $0.13per 1Mvia Alibaba CloudQwen3.7 PlusQwenQwen3.7 Plus is a hosted-only model for text and image questions, with no download to run yourself.writing45th of 168 coding52nd of 168 agents45th of 55 everyday43rd of 168 $0.32 / $1.28per 1Mvia Alibaba CloudQwen3.7 MaxQwenQwen 3.7 Max is a hosted-only text model from Alibaba that can handle up to one million tokens in a single request.writing36th of 168 coding18th of 168 agents40th of 55 everyday24th of 168 $1.25 / $3.75per 1Mvia Novita AIQwen3.65 sizes here
Qwen3.6 FlashQwenQwen3.6 Flash is a proprietary multimodal model from Alibaba that handles up to one million tokens in a single request and accepts text, images and video.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.19 / $1.13per 1Mvia Alibaba CloudQwen3.6 Max PreviewQwenQwen3.6 Max Preview is a hosted-only text model: we list no download for it, so using it means choosing a host.writing38th of 168 coding35th of 168 agentsnot measuredeveryday38th of 168 $1.03 / $6.16per 1Mvia Alibaba CloudQwen3.6 27BQwenAlibaba's 27-billion-parameter mid-size model can be downloaded, released in 2026.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.32 / $2.70per 1Mvia PhalaQwen3.6 35B A3BQwenQwen 3.6 is a multimodal model with a mixture-of-experts design that keeps only 3 billion parameters active for each word while drawing on 36 billion total.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.15 / $1.00per 1Mvia OpenRouterQwen3.6 PlusQwenQwen3.6 Plus is a hosted model with no download, so using it means choosing a host.writing59th of 168 coding55th of 168 agentsnot measuredeveryday54th of 168 $0.33 / $1.95per 1Mvia Alibaba CloudQwen3.53 sizes here
Qwen3.5 Plus 2026-04-20QwenQwen3.5 Plus is Alibaba's proprietary flagship that accepts text, images and video across a one-million-token request limit.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.30 / $1.80per 1Mvia Alibaba CloudQwen3.5 397B A17BQwenQwen 3.5 is a large mixture-of-experts model with 403.4 billion total parameters and 17 billion active per token, released under an Apache licence.writing61st of 168 coding58th of 168 agentsnot measuredeveryday57th of 168 $0.39 / $2.34per 1Mvia Alibaba CloudQwen3.5 Plus 2026-02-15QwenQwen3.5 Plus is a proprietary text model from Alibaba that accepts text, images and video in a single request of up to one million tokens.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.26 / $1.56per 1Mvia Alibaba CloudQwen3.5-9BQwenAlibaba's 9-billion-parameter small downloadable model has a permissive Apache licence and an unusually large request limit for its size.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.10 / $0.15per 1Mvia OpenRouterQwen3.5-FlashQwenQwen3.5-Flash is a hosted-only model: we list no download for it, so using it means choosing a host.writing107th of 168 coding102nd of 168 agentsnot measuredeveryday98th of 168 $0.065 / $0.26per 1Mvia Alibaba CloudQwen3.5-35B-A3BQwenQwen3.5-35B-A3B is a downloadable model you can run yourself, and its licence allows commercial use, changes and redistribution.writing105th of 168 coding104th of 168 agentsnot measuredeveryday101st of 168 $0.16 / $1.30per 1Mvia Alibaba CloudQwen3.5-122B-A10BQwenQwen3.5-122B-A10B is a downloadable model you can run yourself, and its licence allows commercial use, changes and redistribution.writing91st of 168 coding85th of 168 agentsnot measuredeveryday81st of 168 $0.26 / $2.08per 1Mvia Alibaba CloudQwen3.5-27BQwenQwen3.5-27B is a 27.8-billion-parameter multimodal model from Alibaba with a permissive Apache licence and a 262,144-token request limit.writing96th of 168 coding93rd of 168 agentsnot measuredeveryday92nd of 168 $0.20 / $1.56per 1Mvia Alibaba CloudQwen39 sizes here
Qwen3 Max ThinkingQwenQwen3 Max Thinking is a hosted-only reasoning model from Alibaba with a 262,144-token request limit, built for long-document analysis and step-by-step reasoning.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.78 / $3.90per 1Mvia Alibaba CloudQwen3 MaxQwenQwen3 Max is a hosted-only text model: we list no download for it, so using it means choosing a host.writing75th of 168 coding68th of 168 agentsnot measuredeveryday66th of 168 $0.78 / $3.90per 1Mvia Alibaba CloudQwen3 30B A3B Thinking 2507QwenQwen3 is a thinking-mode model from Alibaba with 30.5 billion total parameters and only 3 billion active per word.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.20 / $2.40per 1Mvia Alibaba CloudQwen3 30B A3B Instruct 2507QwenQwen3 is a text-only downloadable model with 30.5 billion total parameters but only 3 billion active per token, released under an Apache licence.writing119th of 168 coding101st of 168 agentsnot measuredeveryday109th of 168 $0.048 / $0.19per 1Mvia OpenRouterQwen3 235B A22B Thinking 2507QwenQwen3 is a large downloadable reasoning model from Alibaba with a permissive Apache licence.writing90th of 168 coding99th of 168 agentsnot measuredeveryday96th of 168 $0.23 / $2.30per 1Mvia DeepInfraQwen3 235B A22B Instruct 2507QwenQwen3 is a large mixture-of-experts model from Alibaba with 235.1 billion total parameters and 22 billion active per token, released in 2025 under an Apache licence.writing86th of 168 coding74th of 168 agentsnot measuredeveryday77th of 168 $0.22 / $0.88per 1Mvia Google Vertex AIQwen3 32BQwenQwen3 is a 32.8-billion-parameter text model released in 2025 under a permissive Apache licence.writing127th of 168 coding122nd of 168 agentsnot measuredeveryday124th of 168 $0.080 / $0.28per 1Mvia OpenRouterQwen3 8BQwenQwen3 is a compact downloadable text model with 8.2 billion parameters and a permissive Apache licence.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.12 / $0.46per 1Mvia Alibaba CloudQwen3 14BQwenQwen3 is a 14.8-billion-parameter text model with a permissive Apache licence and a 131,072-token request limit.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.12 / $0.24per 1Mvia OpenRouterQwen3 Coder NextQwenQwen3 Coder Next is a 79.7-billion-parameter code-specialist model with a permissive Apache licence and a 262,144-token request limit.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.12 / $0.80per 1Mvia OpenRouterQwen3 VL7 sizes here
Qwen3 VL 32B InstructQwenQwen3 VL is a 33.4-billion-parameter vision-language model from Alibaba that accepts text and images under a permissive Apache licence.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.10 / $0.42per 1Mvia Alibaba CloudQwen3 VL 8B InstructQwenQwen3 VL is a compact vision-language model from Alibaba that accepts text and images and returns text.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.12 / $0.46per 1Mvia OpenRouterQwen3 VL 8B ThinkingQwenQwen3 VL is a small downloadable vision-language model that accepts text and images and can show its reasoning step by step.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.18 / $2.10per 1Mvia Alibaba CloudQwen3 VL 30B A3B InstructQwenQwen3 VL is a vision-language model whose weights can be downloaded under a permissive Apache licence.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.15 / $0.60per 1Mvia OpenRouterQwen3 VL 30B A3B ThinkingQwenQwen3 VL is a vision-language model that can reason through problems before answering, with only 3 billion active parameters drawn from 31.1 billion total.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.20 / $2.40per 1Mvia OpenRouterQwen3 VL 235B A22B InstructQwenQwen3 VL 235B A22B Instruct takes text and images in and returns text, and you can download it and run it yourself.writing97th of 168 coding81st of 168 agentsnot measuredeveryday87th of 168 $0.21 / $1.90per 1Mvia OpenRouterQwen3 VL 235B A22B ThinkingQwenQwen3 VL 235B A22B Thinking is a downloadable model that takes text and images and returns text, with a licence that allows commercial use, changes and redistribution (Apache License 2.0). Its measured quality sits mid-field on the boards we hold, so it is a general-purpose pick rather than a leader.writing106th of 168 coding92nd of 168 agentsnot measuredeveryday100th of 168 $0.40 / $4.00per 1Mvia Alibaba CloudQwen3 Coder4 sizes here
Qwen3 Coder PlusQwenQwen3 Coder Plus is a proprietary coding-specialist model from Alibaba with a one-million-token request limit, built for repository-scale tasks.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.65 / $3.25per 1Mvia Alibaba CloudQwen3 Coder FlashQwenQwen3 Coder Flash is a proprietary coding model from Alibaba that can handle up to one million tokens in a single request.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.20 / $0.97per 1Mvia Alibaba CloudQwen3 Coder 30B A3B InstructQwenQwen3 Coder is a downloadable coding specialist with 30.5 billion parameters and only 3 billion active per token, built for software-engineering tasks.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.070 / $0.28per 1Mvia OpenRouterQwen3 Coder 480B A35BQwenQwen3 Coder 480B A35B is a downloadable coding model you can reach through a host, with a licence that allows commercial use, changes and redistribution.writing95th of 168 coding88th of 168 agentsnot measuredeveryday107th of 168 $0.30 / $1.00per 1Mvia OpenRouterQwen3 Next2 sizes here
Qwen3 Next 80B A3B InstructQwenQwen3 Next is a text-only downloadable model from Alibaba with a permissive Apache licence.writing121st of 168 coding98th of 168 agentsnot measuredeveryday97th of 168 $0.10 / $1.10per 1Mvia OpenRouterQwen3 Next 80B A3B ThinkingQwenQwen3 Next is a thinking-mode text model with an Apache licence and a 262,144-token request limit.writing115th of 168 coding114th of 168 agentsnot measuredeveryday113th of 168 $0.15 / $1.20per 1Mvia Google Vertex AIQwen Plus 0728QwenQwen Plus 0728 is a hosted-only text model: we list no download for it, so using it means choosing a host.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.26 / $0.78per 1Mvia Alibaba CloudQwen2.5 VL 72B InstructQwenQwen2.5 VL 72B Instruct reads text and images together and returns written answers, and you can download it to run yourself.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.80 / $1.00per 1Mvia OpenRouterQwen2.5 Coder 32B InstructQwenQwen2.5 Coder is a 32.8-billion-parameter coding specialist you can download and run yourself, released in late 2024 under a permissive Apache licence.writing161st of 168 coding148th of 168 agentsnot measuredeveryday156th of 168 $0.66 / $1.00per 1Mvia Cloudflare Workers AIQwen2.52 sizes here
Qwen2.5 72B InstructQwenQwen 2.5 is a large downloadable text model released in September 2024 with broad Arena coverage across seven categories.writing149th of 168 coding141st of 168 agentsnot measuredeveryday148th of 168 $0.36 / $0.40per 1Mvia OpenRouterQwen2.5 7B InstructQwenQwen 2.5 is a compact downloadable text model with a permissive Apache licence and a 32,768-token request limit.writingnot measuredcodingnot measuredagentsnot measuredeverydaynot measured$0.10 / $0.20per 1Mvia Phala