{"schemaVersion":1,"guide":{"slug":"qwen4-max","title":"Qwen4-Max: everything you need to know","description":"What Qwen4-Max means for work, coding agents and personal projects: sourced benchmark results, task costs, pricing and practical trade-offs.","date":"2026-11-16","path":"/guides/qwen4-max","cover":{"src":"/guides/qwen4-max-cover.png","width":1730,"height":909,"alt":"Notes on Qwen4-Max in a white capsule on a green mesh gradient"}},"sources":{"launch":"https://qwen.ai/blog?id=qwen4-max","api":"https://www.qwencloud.com/models/qwen4-max","independent":"https://artificialanalysis.ai/models/qwen4-max","announcement":"https://x.com/Alibaba_Qwen/status/2122148127520289151","delivery":"https://github.com/QwenLM/Qwen-Live-Delivery-Plugin"},"methodology":{"selection":"Matching independent benchmark version and comparison group; latest dated report, then highest available effort on that date. Effort is not standardized across providers.","missing":"Null means no matching published result; never zero.","tokenCost":"USD per million tokens; equal blend of standard uncached input and output rates. Sources and checkedAt accompany each model."},"benchmarkComparisons":[{"id":"hle-diamond-aa","title":"Reasoning","label":"HLE-Diamond · 1,000 questions · no tools","slug":"hle-diamond","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":44.9,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"No tools · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":80.2,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"No tools · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":58.8,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-sol","settings":"No tools · GPT-6.1 Sol (Max)"},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":47.3,"effort":"max","source":"https://artificialanalysis.ai/evaluations/hle-diamond","settings":"No tools · GPT-6.1 Terra (max)"},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":63.800000000000004,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-astra","settings":"No tools · GPT-6 Astra (Max)"}]},{"id":"gdp-pdf-aa","title":"Document reasoning","label":"GDP.pdf · all-pass · five attempts","slug":"gdp-pdf","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":26.200000000000003,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"No tools · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":24.2,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"No tools · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":31,"effort":"max","source":"https://artificialanalysis.ai/evaluations/gdp-pdf","settings":"No tools · GPT-6.1 Sol (max)"},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":28.6,"effort":"max","source":"https://artificialanalysis.ai/evaluations/gdp-pdf","settings":"No tools · GPT-6.1 Terra (max)"},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":31,"effort":"max","source":"https://artificialanalysis.ai/evaluations/gdp-pdf","settings":"No tools · GPT-6 Astra (max)"}]},{"id":"terminal-bench-4-0-aa","title":"Terminal coding","label":"Terminal-Bench 4.0 · mini-swe-agent · three repeats","slug":"terminal-bench","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":48.4848484848485,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"Tools enabled · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":59.5959595959596,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"Tools enabled · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":56.0606060606061,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-sol","settings":"Tools enabled · GPT-6.1 Sol (Max)"},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":55.6,"effort":"max","source":"https://artificialanalysis.ai/evaluations/terminalbench-4-0","settings":"Tools enabled · GPT-6.1 Terra (max)"},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":59.5959595959596,"effort":"xhigh","source":"https://artificialanalysis.ai/models/gpt-6-astra-xhigh","settings":"Tools enabled · GPT-6 Astra (Xhigh)"}]},{"id":"frontiercode-2-0-aa","title":"Mergeable code","label":"FrontierCode 2.0 · mini-swe-agent · three repeats","slug":"frontiercode","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":26.4534595959596,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"Tools enabled · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":48.7597222222222,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"Tools enabled · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":40.6262878787879,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-sol","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":42.3200505050505,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-terra","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":42.374797979798004,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-astra","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."}]},{"id":"terminal-bench-science-0-1-aa","title":"Scientific workflows","label":"Terminal-Bench-Science 0.1 · independent mini-swe-agent","slug":"terminal-bench-science","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":23.3333333333333,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"Tools enabled · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":73.3333333333333,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"Tools enabled · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":58.0952380952381,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-sol","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":57.619047619047606,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-terra","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":63.3333333333333,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-astra","settings":"Tools enabled · Exact independent Max record; shared mini-swe-agent protocol; three repeats."}]},{"id":"automationbench-aa-1-0-6","title":"Work automations","label":"AutomationBench-AA · v1.0.6 · objective completion","slug":"automationbench-aa","unit":"%","maximum":100,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":72.56119376604481,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"Tools enabled · Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":76.42233832611831,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"Tools enabled · Claude Fable 5.2 (Max, Default Fallback); exact independent record; implementation per archived benchmarking methodology."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":66.6,"effort":"xhigh","source":"https://artificialanalysis.ai/models/gpt-6-1-sol-xhigh","settings":"Tools enabled · GPT-6.1 Sol (xhigh); private held-out 657 tasks; 50 turns; objective completion with zero credit for any guardrail violation. "},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":60.2,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-1-terra","settings":"Tools enabled · GPT-6.1 Terra (max); private held-out 657 tasks; 50 turns; objective completion with zero credit for any guardrail violation. "},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":68.5,"effort":"max","source":"https://artificialanalysis.ai/models/gpt-6-astra","settings":"Tools enabled · GPT-6 Astra (max); private held-out 657 tasks; 50 turns; objective completion with zero credit for any guardrail violation. "}]},{"id":"aa-briefcase-v1-1","title":"Business deliverables","label":"AA-Briefcase v1.1 · Stirrup · combined Elo","slug":"aa-briefcase","unit":"Elo","maximum":1800,"models":[{"id":"qwen4-max","name":"Qwen4-Max","color":"#d6ff62","score":1686.07,"effort":"xhigh","source":"https://artificialanalysis.ai/models/qwen4-max","settings":"Tools enabled · Exact independent xhigh record. Combined Elo, not rubric pass percentage."},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","color":"#ee7956","score":1785.32,"effort":"max","source":"https://artificialanalysis.ai/models/claude-fable-5-2","settings":"Tools enabled · Claude Fable 5.2 (Max, Default Fallback); combined Elo; not rubric percentage."},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","color":"#9fc6fb","score":1555,"effort":"max","source":"https://artificialanalysis.ai/evaluations/aa-briefcase","settings":"Tools enabled · GPT-6.1 Sol (max)"},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","color":"#88d9c1","score":1402,"effort":"max","source":"https://artificialanalysis.ai/evaluations/aa-briefcase","settings":"Tools enabled · GPT-6.1 Terra (max)"},{"id":"gpt-6-astra","name":"GPT-6 Astra","color":"#b9b3ff","score":1569,"effort":"max","source":"https://artificialanalysis.ai/evaluations/aa-briefcase","settings":"Tools enabled · GPT-6 Astra (max)"}]}],"launchFindings":[{"slug":"work360","title":"A cheaper first pass at professional work.","benchmark":"Work360 · internal evaluation","score":52.69,"previous":47.74,"unit":"score","commentary":"Think about a client brief that needs a spreadsheet, a written analysis and a presentation. Qwen’s Work360 framework covers 26 broad industry domains, with examples ranging from campaign reviews to engineering handoffs. Its reported score rises by 4.95 points over Qwen3.8-Max. That gives a small team another model to test on the production work around a project: gathering evidence, reconciling files and building a first deliverable.","detail":"Work360 is Qwen’s internal evaluation. 52.69 is its reported score; the launch does not establish that 52.69% of arbitrary client projects will be completed successfully.","source":"https://qwen.ai/blog?id=qwen4-max#work360"},{"slug":"agents-last-exam","title":"Long jobs need checkpoints you can review.","benchmark":"Agents’ Last Exam v1.1 · launch snapshot","score":46.7,"previous":27.9,"commentary":"The 18.8-point increase on this long-horizon task evaluation is one of the larger changes in the launch chart. It makes multi-step assignments worth testing: review a folder, identify the important issues, produce a document, then check it against the brief. Qwen’s Live Delivery approach returns usable versions along the way so the person directing the work can correct the course early.","detail":"Launch-chart scores. The exact task release, agent harness, effort and time budget are not specified in the article’s footnotes. Kept separate from Google’s October-task-set track.","source":"https://qwen.ai/blog?id=qwen4-max"},{"slug":"deepswe","title":"Coding gains are uneven across tests.","benchmark":"DeepSWE v1.1 · launch snapshot","score":75.7,"previous":69.3,"commentary":"Qwen4-Max improves by 6.4 points on DeepSWE. That is encouraging for repository work, but the same launch chart puts it at 31.5 on ProgramBench, compared with 65.0 for Opus 5.5. The practical move is to test it on the kind of engineering your agent actually does: fixing bugs, implementing features or reproducing an existing application. A strong result in one category does not settle the others.","detail":"Qwen’s launch comparison. DeepSWE and ProgramBench measure different tasks; these figures are not developer productivity estimates. The prior-model label in the chart is Qwen3.8-Max-0902.","source":"https://qwen.ai/blog?id=qwen4-max"},{"slug":"automationbench-aa","title":"Useful workflow progress still needs an end-to-end check.","benchmark":"AutomationBench-AA · independent xhigh","score":72.56119376604481,"commentary":"Artificial Analysis reports about 72.6% on objective-level workflow scoring at xhigh. For a business, that suggests testing sequences such as finding a record, applying a policy and preparing the next step. A partly correct workflow can still leave the customer waiting, so your own evaluation should check whether the entire job finished and whether any actions were wrong.","detail":"AA’s private 657-task, v1.0.6, 50-turn protocol. Objective completion without guardrail violations is distinct from Zapier’s strict task-completion metric.","source":"https://artificialanalysis.ai/models/qwen4-max"},{"slug":"aa-briefcase","title":"Judge the files it hands back.","benchmark":"AA-Briefcase v1.1 · independent xhigh","score":1686.07,"unit":"Elo","commentary":"AA-Briefcase evaluates complex knowledge work and the quality of the resulting deliverables. Qwen4-Max’s combined rating is 1686.07 Elo in the independent record. This is relevant when your output is a workbook or presentation someone else must use. Give it source files and a clear rubric, then inspect the calculations, analytical quality and presentation before handing the work to a client.","detail":"Combined Elo is a relative quality rating, not percentage accuracy or the share of finished projects. Qwen’s launch rounds this rating to 1686.","source":"https://artificialanalysis.ai/models/qwen4-max"},{"slug":"chartography","title":"Native vision helps with documents, but check the numbers.","benchmark":"Chartography · Qwen image-resolution setup","score":54.2,"previous":42,"commentary":"The launch reports a 12.2-point improvement in chart understanding. That is useful for work that starts with a deck or a visual report: read a chart, identify a trend and turn it into an explanation. Keep the original data nearby. A chart-reading benchmark does not prove the model will reliably extract every value from a client’s slide.","detail":"Qwen sets min_pixels to 4096×32×32. This is a separate protocol from the Anthropic implementation and its grader. Tool access is unspecified.","source":"https://qwen.ai/blog?id=qwen4-max#footnotes"}],"tokenCosts":{"models":[{"id":"deepseek-v4-1-flash-1017","name":"DeepSeek V4.1 Flash 1017","provider":"DeepSeek","logo":"/providers/deepseek.svg","isOpen":1,"cost":0.196,"input":0.098,"output":0.294,"host":"DeepInfra","offeringId":"aa:70bb5da0-0732-4130-b626-20165376ff80","configuration":"DeepSeek V4.1 Flash 1017 (max)","source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash-1017/providers","checkedAt":"2026-11-02T21:05:33.228000+00:00"},{"id":"deepseek-v4-1-flash","name":"DeepSeek V4.1 Flash","provider":"DeepSeek","logo":"/providers/deepseek.svg","isOpen":1,"cost":0.28,"input":0.14,"output":0.42,"host":"DeepInfra","offeringId":"aa:077ec157-ab1f-48d9-a7cb-b3678f512a87","configuration":"DeepSeek V4.1 Flash (max)","source":"https://artificialanalysis.ai/models/deepseek-v4-1-flash/providers","checkedAt":"2026-11-02T21:05:23.617000+00:00"},{"id":"claude-haiku-5-5","name":"Claude Haiku 5.5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":0.3,"input":0.1,"output":0.5,"host":"Anthropic","offeringId":"coverage-claude-haiku-5-5-anthropic-87e5e4f6","configuration":"Standard API, prompts up to 100K input tokens","source":"https://platform.claude.com/docs/en/models/haiku-5-5/overview","checkedAt":"2026-11-02T23:00:00Z"},{"id":"gpt-6-luna","name":"GPT-6 Luna","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":0.3,"input":0.1,"output":0.5,"host":"OpenAI","offeringId":"coverage-gpt-6-luna-openai-0ce70a3f","configuration":"Standard API, prompts up to 272K input tokens","source":"https://developers.openai.com/api/docs/models/gpt-6-luna","checkedAt":"2026-11-02T23:00:00Z"},{"id":"glm-5-3-flash","name":"GLM-5.3-Flash","provider":"Z.ai","logo":"/providers/zai.svg","isOpen":1,"cost":0.325,"input":0.15,"output":0.5,"host":"Z.ai","offeringId":"open-import-glm-5-3-flash-zai","configuration":"GLM-5.3-Flash","source":"https://artificialanalysis.ai/models/glm-5-3-flash","checkedAt":"2026-11-02T22:18:13.301725+00:00"},{"id":"minimax-m3-1-flash","name":"MiniMax M3.1 Flash","provider":"MiniMax","logo":"/providers/minimax.ico","isOpen":1,"cost":0.35624999999999996,"input":0.1425,"output":0.57,"host":"GMI","offeringId":"aa:8e90a589-a404-40fc-bf52-0920031859c9","configuration":"MiniMax-M3.1-Flash","source":"https://artificialanalysis.ai/models/minimax-m3-1-flash/providers","checkedAt":"2026-11-02T21:06:38.392000+00:00"},{"id":"glm-5-3","name":"GLM-5.3","provider":"Z.ai","logo":"/providers/zai.svg","isOpen":1,"cost":0.42000000000000004,"input":0.2,"output":0.64,"host":"Multiverse Computing","offeringId":"aa:daefee47-17f9-45c4-b0a4-f4ba0212cfc4","configuration":"GLM-5.3 (max)","source":"https://artificialanalysis.ai/models/glm-5-3/providers","checkedAt":"2026-11-02T21:06:11.906000+00:00"},{"id":"mimo-v2-6-pro","name":"MiMo-V2.6-Pro","provider":"Xiaomi","logo":"/providers/xiaomi.svg","isOpen":1,"cost":0.6525,"input":0.435,"output":0.87,"host":"Xiaomi","offeringId":"coverage-mimo-v2-6-pro-xiaomi-d7b70af7","configuration":"Standard API, USD, uncached input","source":"https://mimo.mi.com/models/en-US/mimo-v2-6-pro","checkedAt":"2026-11-02T23:00:00Z"},{"id":"minimax-m3","name":"MiniMax M3","provider":"MiniMax","logo":"/providers/minimax.ico","isOpen":1,"cost":0.75,"input":0.3,"output":1.2,"host":"MiniMax","offeringId":"open-import-minimax-m3-minimax","configuration":"MiniMax-M3","source":"https://artificialanalysis.ai/models/minimax-m3","checkedAt":"2026-11-02T22:18:13.301725+00:00"},{"id":"deepseek-v4-pro-0813","name":"DeepSeek V4 Pro 0813","provider":"DeepSeek","logo":"/providers/deepseek.svg","isOpen":1,"cost":1.25,"input":0.5,"output":2,"host":"Systalyze","offeringId":"aa:877c8764-fedb-44b6-aaa8-d7a26ae9892f","configuration":"DeepSeek V4 Pro 0813 (max)","source":"https://artificialanalysis.ai/models/deepseek-v4-pro/providers","checkedAt":"2026-11-02T21:06:38.358000+00:00"},{"id":"gemini-3-8-flash","name":"Gemini 3.8 Flash","provider":"Google","logo":"/providers/google.svg","isOpen":0,"cost":2.25,"input":0.75,"output":3.75,"host":"Google AI Studio","offeringId":"aa:9d7c85a8-a637-4dec-8710-d04e252fddd2","configuration":"Gemini 3.8 Flash (high) AI Studio","source":"https://artificialanalysis.ai/models/gemini-3-8-flash/providers","checkedAt":"2026-11-02T21:06:11.263000+00:00"},{"id":"kimi-k2-6","name":"Kimi K2.6","provider":"Moonshot AI","logo":"/providers/moonshot-ai.svg","isOpen":1,"cost":2.475,"input":0.95,"output":4,"host":"Moonshot AI","offeringId":"open-import-kimi-k2-6-moonshot-ai","configuration":"Kimi K2.6 (Reasoning)","source":"https://artificialanalysis.ai/models/kimi-k2-6","checkedAt":"2026-11-02T22:18:13.301725+00:00"},{"id":"inkling","name":"Inkling","provider":"Thinking Machines","logo":"/providers/thinking-machines.svg","isOpen":1,"cost":2.525,"input":1,"output":4.05,"host":"Thinking Machines","offeringId":"coverage-inkling-thinking-machines-1f42c605","configuration":"Serverless inference beta, 256K context","source":"https://thinkingmachines.ai/tinker/pricing/","checkedAt":"2026-11-02T23:00:00Z"},{"id":"muse-spark-1-1","name":"Muse Spark 1.1","provider":"Meta","logo":"/providers/meta.svg","isOpen":0,"cost":2.75,"input":1.25,"output":4.25,"host":"Meta","offeringId":"aa:f56ad4b4-8302-4a4a-8873-75d70eec2fd8","configuration":"Muse Spark 1.1 (xhigh)","source":"https://artificialanalysis.ai/models/muse-spark-1-1/providers","checkedAt":"2026-11-02T21:06:38.213000+00:00"},{"id":"muse-spark-1-2","name":"Muse Spark 1.2","provider":"Meta","logo":"/providers/meta.svg","isOpen":0,"cost":2.75,"input":1.25,"output":4.25,"host":"Meta","offeringId":"aa:650f6825-32b2-4486-bee7-cd58ff5462f4","configuration":"Muse Spark 1.2 (xhigh)","source":"https://artificialanalysis.ai/models/muse-spark-1-2/providers","checkedAt":"2026-11-02T21:06:38.386000+00:00"},{"id":"muse-spark-1-3","name":"Muse Spark 1.3","provider":"Meta","logo":"/providers/meta.svg","isOpen":0,"cost":2.75,"input":1.25,"output":4.25,"host":"Meta","offeringId":"aa:0d5ad9ff-4812-4379-bad4-b0393a5a80e4","configuration":"Muse Spark 1.3 (max)","source":"https://artificialanalysis.ai/models/muse-spark-1-3/providers","checkedAt":"2026-11-02T21:06:37.992000+00:00"},{"id":"glm-5-2","name":"GLM-5.2","provider":"Z.ai","logo":"/providers/zai.svg","isOpen":1,"cost":2.9000000000000004,"input":1.4,"output":4.4,"host":"Z.ai","offeringId":"open-import-glm-5-2-zai","configuration":"GLM-5.2 (Max)","source":"https://artificialanalysis.ai/models/glm-5-2","checkedAt":"2026-11-02T22:18:13.301725+00:00"},{"id":"gpt-6-1-terra","name":"GPT-6.1 Terra","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":3,"input":1,"output":5,"host":"OpenAI","offeringId":"coverage-gpt-6-1-terra-openai-0ce70a3f","configuration":"Standard API, prompts up to 272K input tokens","source":"https://developers.openai.com/api/docs/models/gpt-6.1-terra","checkedAt":"2026-11-02T23:00:00Z"},{"id":"qwen4-max","name":"Qwen4-Max","provider":"Qwen","logo":"/providers/qwen.jpg","isOpen":0,"cost":3,"input":1.5,"output":4.5,"host":"QwenCloud","offeringId":"official:qwen4-max:standard","configuration":"Standard API, Singapore; uncached input","source":"https://www.qwencloud.com/models/qwen4-max","checkedAt":"2026-11-16T00:00:00Z"},{"id":"grok-4-7","name":"Grok 4.7","provider":"xAI","logo":"/providers/xai.svg","isOpen":0,"cost":4,"input":2,"output":6,"host":"SpaceXAI","offeringId":"aa:d7d9eaf3-1807-4d16-b5b6-c2b21897cad7","configuration":"Grok 4.7 (xhigh)","source":"https://artificialanalysis.ai/models/grok-4-7/providers","checkedAt":"2026-11-02T21:06:12.238000+00:00"},{"id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","provider":"Google","logo":"/providers/google.svg","isOpen":0,"cost":4.5,"input":1.5,"output":7.5,"host":"Google","offeringId":"official:gemini-4-hydrogen:standard","configuration":"Standard Gemini API; independent speed measured at xhigh","source":"https://aistudio.google.com/docs/pricing#gemini-4-hydrogen","checkedAt":"2026-11-12T19:17:51.391706+00:00"},{"id":"claude-sonnet-5","name":"Claude Sonnet 5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":6,"input":2,"output":10,"host":"Amazon","offeringId":"aa:2beeef6c-2a88-4b90-8055-90070b81943d","configuration":"Claude Sonnet 5 (max)","source":"https://artificialanalysis.ai/models/claude-sonnet-5/providers","checkedAt":"2026-11-02T21:05:33.716000+00:00"},{"id":"claude-sonnet-5-5","name":"Claude Sonnet 5.5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":6,"input":2,"output":10,"host":"Amazon","offeringId":"aa:24eed6aa-384b-44cc-bb60-a3d58da0a52a","configuration":"Claude Sonnet 5.5 (max with fallback)","source":"https://artificialanalysis.ai/models/claude-sonnet-5-5/providers","checkedAt":"2026-11-02T21:05:33.869000+00:00"},{"id":"gpt-6-1-sol","name":"GPT-6.1 Sol","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":6,"input":2,"output":10,"host":"Azure","offeringId":"aa:5d98a9a9-5178-4783-a2a4-271a944a8bf2","configuration":"GPT-6.1 Sol (max)","source":"https://artificialanalysis.ai/models/gpt-6-1-sol/providers","checkedAt":"2026-11-02T21:06:11.475000+00:00"},{"id":"kimi-k3-1-pro","name":"Kimi K3.1 Pro","provider":"Moonshot AI","logo":"/providers/moonshot-ai.svg","isOpen":1,"cost":6,"input":2,"output":10,"host":"Kimi","offeringId":"aa:bdb8515f-f143-43c0-9b0d-60048baabec5","configuration":"Kimi K3.1 Pro","source":"https://artificialanalysis.ai/models/kimi-k3-1-pro/providers","checkedAt":"2026-11-02T21:06:38.811000+00:00"},{"id":"gemini-3-1-pro","name":"Gemini 3.1 Pro","provider":"Google","logo":"/providers/google.svg","isOpen":0,"cost":7,"input":2,"output":12,"host":"Google (AI Studio)","offeringId":"aa:6818e291-75eb-43eb-9c77-5c455f42d5da","configuration":"Gemini 3.1 Pro Preview (AI Studio)","source":"https://artificialanalysis.ai/models/gemini-3-1-pro-preview/providers","checkedAt":"2026-11-02T21:06:37.869000+00:00"},{"id":"gpt-5-6-terra","name":"GPT-5.6 Terra","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":7,"input":2,"output":12,"host":"OpenAI","offeringId":"coverage-gpt-5-6-terra-openai-0ce70a3f","configuration":"Standard API, prompts up to 272K input tokens","source":"https://developers.openai.com/api/docs/models/gpt-5.6-terra","checkedAt":"2026-11-02T23:00:00Z"},{"id":"kimi-k3","name":"Kimi K3","provider":"Moonshot AI","logo":"/providers/moonshot-ai.svg","isOpen":1,"cost":7.825,"input":0.65,"output":15,"host":"Makora","offeringId":"aa:267b5a3f-0203-4378-92a2-20e05a41116b","configuration":"Kimi K3 (max)","source":"https://artificialanalysis.ai/models/kimi-k3/providers","checkedAt":"2026-11-02T21:06:11.860000+00:00"},{"id":"claude-opus-5-5","name":"Claude Opus 5.5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":12,"input":4,"output":20,"host":"Amazon","offeringId":"aa:032a278a-7f1b-4294-afd8-e64e1486c88e","configuration":"Claude Opus 5.5 (max with fallback)","source":"https://artificialanalysis.ai/models/claude-opus-5-5/providers","checkedAt":"2026-11-02T21:05:32.192000+00:00"},{"id":"gemini-4-argon","name":"Gemini 4 Argon","provider":"Google","logo":"/providers/google.svg","isOpen":0,"cost":12,"input":4,"output":20,"host":"Google (Vertex)","offeringId":"aa:59c14390-24f2-412e-9c05-6d11952d2d9c","configuration":"Gemini 4 Argon (high) (Vertex)","source":"https://artificialanalysis.ai/models/gemini-4-argon/providers","checkedAt":"2026-11-02T21:06:12.205000+00:00"},{"id":"gemini-4-carbon","name":"Gemini 4 Carbon","provider":"Google","logo":"/providers/google.svg","isOpen":0,"cost":12,"input":4,"output":20,"host":"Google","offeringId":"aa:c23d5fdc-52b0-42a7-af0c-449089896e22","configuration":"Gemini 4 Carbon (high)","source":"https://artificialanalysis.ai/models/gemini-4-carbon/providers","checkedAt":"2026-11-02T21:06:10.181000+00:00"},{"id":"gpt-5-6-sol","name":"GPT-5.6 Sol","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":12,"input":4,"output":20,"host":"Amazon","offeringId":"aa:ddd0a4c4-a366-4f8e-bda2-874ece77dbf3","configuration":"GPT-5.6 Sol (max)","source":"https://artificialanalysis.ai/models/gpt-5-6-sol/providers","checkedAt":"2026-11-02T21:06:37.764000+00:00"},{"id":"muse-wonder-1-0","name":"Muse Wonder 1.0","provider":"Meta","logo":"/providers/meta.svg","isOpen":0,"cost":12,"input":4,"output":20,"host":"Meta","offeringId":"coverage-muse-wonder-1-0-meta-5470ef52","configuration":"Standard API, uncached input","source":"https://dev.meta.ai/resources/blog/build-with-muse-wonder","checkedAt":"2026-11-02T23:00:00Z"},{"id":"claude-opus-4-6","name":"Claude Opus 4.6","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":15,"input":5,"output":25,"host":"Amazon","offeringId":"aa:fcc25cee-9e96-44ab-bf62-7eec8888f44e","configuration":"Claude Opus 4.6 (non-reasoning, high)","source":"https://artificialanalysis.ai/models/claude-opus-4-6/providers","checkedAt":"2026-11-02T21:05:33.492000+00:00"},{"id":"claude-opus-4-7","name":"Claude Opus 4.7","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":15,"input":5,"output":25,"host":"Amazon","offeringId":"aa:bfe8e92a-ee64-468a-bb9e-472bf7810884","configuration":"Claude Opus 4.7 (max)","source":"https://artificialanalysis.ai/models/claude-opus-4-7/providers","checkedAt":"2026-11-02T21:05:33.936000+00:00"},{"id":"claude-opus-4-8","name":"Claude Opus 4.8","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":15,"input":5,"output":25,"host":"Amazon","offeringId":"aa:c2dfbde3-60dd-4b88-b0c0-7b7ef680e933","configuration":"Claude Opus 4.8 (max)","source":"https://artificialanalysis.ai/models/claude-opus-4-8/providers","checkedAt":"2026-11-02T21:05:33.666000+00:00"},{"id":"claude-opus-5","name":"Claude Opus 5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":15,"input":5,"output":25,"host":"Amazon","offeringId":"aa:3bf58af8-d004-4ad0-8fbc-81615b088d4f","configuration":"Claude Opus 5 (max)","source":"https://artificialanalysis.ai/models/claude-opus-5/providers","checkedAt":"2026-11-02T21:05:33.997000+00:00"},{"id":"gpt-5-5","name":"GPT-5.5","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":17.5,"input":5,"output":30,"host":"OpenAI","offeringId":"aa:65aec0e6-b670-443a-81c2-79e38c4d365a","configuration":"GPT-5.5 (xhigh)","source":"https://artificialanalysis.ai/models/gpt-5-5/providers","checkedAt":"2026-11-02T21:06:12.002000+00:00"},{"id":"claude-fable-5-2","name":"Claude Fable 5.2","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":24,"input":8,"output":40,"host":"Anthropic","offeringId":"official:claude-fable-5-2:standard","configuration":"Standard uncached API","source":"https://platform.claude.com/docs/en/about-claude/pricing","checkedAt":"2026-11-09T00:00:00Z"},{"id":"claude-mythos-5-2","name":"Claude Mythos 5.2","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":24,"input":8,"output":40,"host":"Anthropic","offeringId":"official:claude-mythos-5-2:standard","configuration":"Standard uncached API","source":"https://platform.claude.com/docs/en/about-claude/pricing","checkedAt":"2026-11-09T00:00:00Z"},{"id":"claude-fable-5","name":"Claude Fable 5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":30,"input":10,"output":50,"host":"Anthropic","offeringId":"aa:c585177d-d38e-498b-a0d6-3b9c20fdedd3","configuration":"Claude Fable 5 (with fallback)","source":"https://artificialanalysis.ai/models/claude-fable-5/providers","checkedAt":"2026-11-02T21:05:32.564000+00:00"},{"id":"claude-fable-5-1","name":"Claude Fable 5.1","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":30,"input":10,"output":50,"host":"Anthropic","offeringId":"aa:c7e37b8e-9e3c-4bc2-a055-1f6db77cdad1","configuration":"Claude Fable 5.1 (max with fallback)","source":"https://artificialanalysis.ai/models/claude-fable-5-1/providers","checkedAt":"2026-11-02T21:05:32.970000+00:00"},{"id":"claude-mythos-5","name":"Claude Mythos 5","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":30,"input":10,"output":50,"host":"Anthropic","offeringId":"official:claude-mythos-5","configuration":"Official standard API pricing","source":"https://platform.claude.com/docs/en/about-claude/pricing","checkedAt":"2026-11-02T00:00:00+00:00"},{"id":"gpt-6-astra","name":"GPT-6 Astra","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":30,"input":10,"output":50,"host":"Amazon","offeringId":"aa:568239b4-c32b-4d68-89e8-9c01db3b846f","configuration":"GPT-6 Astra (max)","source":"https://artificialanalysis.ai/models/gpt-6-astra/providers","checkedAt":"2026-11-02T21:06:11.700000+00:00"},{"id":"claude-mythos-preview","name":"Claude Mythos Preview","provider":"Anthropic","logo":"/providers/anthropic.svg","isOpen":0,"cost":75,"input":25,"output":125,"host":"Anthropic","offeringId":"official:claude-mythos-preview","configuration":"Official standard API pricing","source":"https://www.anthropic.com/glasswing","checkedAt":"2026-11-02T00:00:00+00:00"},{"id":"gpt-5-4-pro","name":"GPT-5.4 Pro","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":105,"input":30,"output":180,"host":"OpenAI","offeringId":"aa:ac8f47f7-d6f0-4890-9456-73a44df31d05","configuration":"GPT-5.4 Pro (xhigh)","source":"https://artificialanalysis.ai/models/gpt-5-4-pro/providers","checkedAt":"2026-11-02T21:06:10.856000+00:00"},{"id":"gpt-5-5-pro","name":"GPT-5.5 Pro","provider":"OpenAI","logo":"/providers/openai.svg","isOpen":0,"cost":105,"input":30,"output":180,"host":"OpenAI","offeringId":"official:gpt-5-5-pro","configuration":"Official standard API pricing","source":"https://openai.com/index/introducing-gpt-5-5/","checkedAt":"2026-11-02T00:00:00+00:00"}],"total":99},"work360TaskCost":{"source":"https://qwen.ai/blog?id=qwen4-max","metric":"Average USD per attempted task in Qwen's internal evaluation","models":[{"model":"Qwen3.8-Max","score":47.74,"cost":4.43},{"model":"Qwen4-Max","score":52.69,"cost":1.67}]},"agentIndex":{"model":{"modelId":"qwen4-max","model":"Qwen4-Max","provider":"Qwen","score":null,"rank":null,"coverage":4,"status":"insufficient_coverage"},"components":[{"modelId":"qwen4-max","key":"reasoning","label":"HLE-Diamond","benchmarkId":"hle-diamond-aa","benchmarkVersion":"1,000 questions / no tools","fallback":0,"weight":0.1,"anchorMin":18.2,"anchorMax":67.9,"resultId":1539,"rawScore":44.9,"normalizedScore":53.72233400402414,"contribution":5.372233400402415,"source":"https://artificialanalysis.ai/models/qwen4-max","settings":"{\"comparison_group\": \"aa-hle-diamond-no-tools-1000\", \"effort\": \"xhigh\", \"settings\": \"Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh.\", \"tools\": false}"},{"modelId":"qwen4-max","key":"documents","label":"GDP.pdf","benchmarkId":"gdp-pdf-aa","benchmarkVersion":"100 tasks / All-pass / five attempts","fallback":0,"weight":0.1,"anchorMin":15.4,"anchorMax":31.4,"resultId":1540,"rawScore":26.200000000000003,"normalizedScore":67.50000000000003,"contribution":6.7500000000000036,"source":"https://artificialanalysis.ai/models/qwen4-max","settings":"{\"comparison_group\": \"aa-gdp-pdf-all-pass-five-attempts\", \"effort\": \"xhigh\", \"settings\": \"Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh.\", \"tools\": false}"},{"modelId":"qwen4-max","key":"deliverables","label":"AA-Briefcase","benchmarkId":"aa-briefcase-v1-1","benchmarkVersion":"1.1 / combined Elo","fallback":0,"weight":0.2,"anchorMin":1203,"anchorMax":1814,"resultId":1545,"rawScore":1686.07,"normalizedScore":79.0621931260229,"contribution":15.812438625204582,"source":"https://artificialanalysis.ai/models/qwen4-max","settings":"{\"comparison_group\": \"aa-briefcase-1.1-stirrup-combined-elo\", \"effort\": \"xhigh\", \"settings\": \"Exact independent xhigh record. Combined Elo, not rubric pass percentage.\", \"tools\": true}"},{"modelId":"qwen4-max","key":"agentic","label":"LiveBench Agentic","benchmarkId":"livebench-2-agentic-2026-11-06","benchmarkVersion":null,"fallback":0,"weight":0.2,"anchorMin":16.5,"anchorMax":77.9,"resultId":null,"rawScore":null,"normalizedScore":null,"contribution":null,"source":null,"settings":null},{"modelId":"qwen4-max","key":"coding","label":"LiveBench Coding","benchmarkId":"livebench-2-coding-2026-11-06","benchmarkVersion":null,"fallback":0,"weight":0.05,"anchorMin":59.3,"anchorMax":97.7,"resultId":null,"rawScore":null,"normalizedScore":null,"contribution":null,"source":null,"settings":null},{"modelId":"qwen4-max","key":"terminal","label":"Terminal-Bench 4.0","benchmarkId":"terminal-bench-4-0-aa","benchmarkVersion":"4.0 / mini-swe-agent","fallback":0,"weight":0.15,"anchorMin":12.6262626262626,"anchorMax":67.1717171717172,"resultId":1541,"rawScore":48.4848484848485,"normalizedScore":65.74074074074075,"contribution":9.861111111111112,"source":"https://artificialanalysis.ai/models/qwen4-max","settings":"{\"comparison_group\": \"aa-terminal-4.0-mini-swe-agent-three-repeats\", \"effort\": \"xhigh\", \"settings\": \"Exact independent xhigh record; protocol per archived Artificial Analysis methodology. Effort: xhigh.\", \"tools\": true}"},{"modelId":"qwen4-max","key":"browser","label":"LiveBench Browser Use","benchmarkId":"livebench-2-browser-use-2026-11-06","benchmarkVersion":null,"fallback":0,"weight":0.2,"anchorMin":45.1,"anchorMax":93.3,"resultId":null,"rawScore":null,"normalizedScore":null,"contribution":null,"source":null,"settings":null}]}}