# Qwen3.8 27B Qwen3.8 27B is a text generation model from Qwen, available through the Impossibl API. Its published context window is 1,048,576 tokens. The advertised input rate is $0.50 per million input tokens. The advertised output rate is $3.00 per million output tokens. - Model identifier: ` qwen/qwen3.8-27b ` - Creator: Qwen - Model type: Text generation - Canonical model page: [Qwen3.8 27B]() - API availability: Listed as serving through Impossibl. - Creator release date: 2026-08-14 ## Capabilities and limits - Context window: 1,048,576 tokens - Maximum output tokens: Not published in the public catalog; this is separate from the context window. - Inputs: text, image - Outputs: text - Tool calling: Not published ## API access Use ` qwen/qwen3.8-27b ` as the model identifier through the Impossibl API. Published API endpoints: - ` /v1/chat/completions ` [API quickstart]() · [Models and billing documentation]() ## Published API prices All rates are in USD. An unpublished rate is unknown, not zero. Workspace-specific pricing, taxes, balance-purchase fees, billing adjustments, and tool charges can affect the amount paid. | Usage | Published base rate | Billing unit | | --- | --- | --- | | Input | $0.50 | per million input tokens | | Output | $3.00 | per million output tokens | | Cache read | $0.10 | per million cached input tokens | | Cache write (default retention) | Not published | per million tokens | | Cache write (one-hour retention) | Not published | per million tokens | | Audio input | Not published | per million audio input tokens | ## Independent benchmarks Source: [Artificial Analysis](). Last checked: 2026-09-11T12:23:22.823Z. Configurations are evaluated separately. API scores retain their original units. Unmeasured scores are omitted. ### Qwen3.8 27B (xhigh) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 33.90 | | Omniscience Index | -9.98 | | GPQA | 90.51% | | Humanity’s Last Exam | 33.92% | | SciCode | 46.64% | | CritPt | 5.43% | | Terminal-Bench v2.1 | 79.78% | | Terminal-Bench v4.0 | 5.56% | | AA-LCR | 82.00% | | Multilingual LCR | 21.67% | | GDPval Elo | 1463.25 | | GDP.pdf All-pass | 16.40% | | AutomationBench Partial | 48.24% | | MMMU-Pro | 76.30% | | Tau Banking | 48.04% | | coding index (API) | 68.1 | | agentic index (API) | 46.5 | | intelligence index (API) | 33.9 | ### Qwen3.8 27B (low) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 26.49 | | Omniscience Index | -26.67 | | GPQA | 84.55% | | Humanity’s Last Exam | 14.04% | | SciCode | 40.05% | | CritPt | 0.00% | | Terminal-Bench v2.1 | 67.42% | | Terminal-Bench v4.0 | 2.53% | | AA-LCR | 77.33% | | GDPval Elo | 1402.54 | | GDP.pdf All-pass | 10.40% | | AutomationBench Partial | 30.10% | | MMMU-Pro | 73.76% | | Tau Banking | 32.16% | | coding index (API) | 58.2 | | agentic index (API) | 39.9 | | intelligence index (API) | 26.5 | ### Qwen3.8 27B (medium) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 27.81 | | Omniscience Index | -36.15 | | GPQA | 84.55% | | Humanity’s Last Exam | 14.09% | | SciCode | 39.00% | | CritPt | 0.00% | | Terminal-Bench v2.1 | 65.17% | | Terminal-Bench v4.0 | 5.05% | | AA-LCR | 79.67% | | GDPval Elo | 1430.39 | | GDP.pdf All-pass | 12.80% | | AutomationBench Partial | 40.25% | | MMMU-Pro | 74.16% | | Tau Banking | 47.42% | | coding index (API) | 56.1 | | agentic index (API) | 45.3 | | intelligence index (API) | 27.8 | ### Qwen3.8 27B (Non-reasoning) Source: [Evaluation source]() | Benchmark | Score | | --- | --- | | Intelligence Index | 22.42 | | Omniscience Index | -7.95 | | GPQA | 81.82% | | Humanity’s Last Exam | 12.14% | | SciCode | 36.23% | | CritPt | 0.29% | | Terminal-Bench v2.1 | 49.06% | | AA-LCR | 69.33% | | GDPval Elo | 1136.81 | | MMMU-Pro | 69.94% | | Tau Banking | 20.00% | | coding index (API) | 44.6 | | intelligence index (API) | 22.4 | Full benchmark data and breakdowns: [Complete benchmark snapshot](). ## Availability and retirement The model is listed as serving. No retirement is announced in the current public catalog. ## Frequently asked questions ### What is Qwen3.8 27B? Qwen3.8 27B is a text-generation model from Qwen, available through the Impossibl API. ### How much does Qwen3.8 27B cost? Qwen3.8 27B costs $0.50 per million input tokens and $3.00 per million output tokens through the Impossibl API. Cache reads cost $0.10 per million tokens. All rates are in USD. ### What is the context length of Qwen3.8 27B? Qwen3.8 27B has a 1,048,576-token context window through the Impossibl API. ### What inputs and outputs does Qwen3.8 27B support? Qwen3.8 27B accepts text and image as input. Qwen3.8 27B returns text. ### How do I use Qwen3.8 27B through an API? Qwen3.8 27B is available through the Impossibl API. Use qwen/qwen3.8-27b as the model identifier with /v1/chat/completions. The quickstart explains how to create an API key and send your first request. ### When was Qwen3.8 27B released? Qwen released Qwen3.8 27B on August 14, 2026. Sources: [Creator release announcement]() ## Sources and related links Prices, limits, capabilities, endpoints, and availability come from the public Impossibl catalog. Creator documentation supplies descriptions and release dates where available; it does not override the published API offering. - [Public Impossibl model catalog]() - [Full model directory as JSON]() - [Qwen3.8 27B: model details and pricing]() - [Model directory]()