How much does Qwen 3.8 27B cost?

Qwen 3.8 27B costs $0.45 per million input tokens and $3.00 per million output tokens, using the median price across hosting providers as of 26 Oct 2026. That is about $0.20 for 100 chats of about 1,000 tokens in and 500 tokens back. Confirm on the maker's site before you buy, because prices change.

What is the context window of Qwen 3.8 27B?

Qwen 3.8 27B has a context window of 262K tokens, roughly 197,000 words of text read at once. A single answer can be up to 33K tokens long.

Is Qwen 3.8 27B open source?

Qwen 3.8 27B has open weights: its trained parameters can be downloaded. That is not always the same as open source, because licence terms differ, so read the licence before you use it commercially.

How good is Qwen 3.8 27B compared with other AI models?

On the Epoch Capabilities Index, Qwen 3.8 27B scores 151.7 (likely range 149.8 to 153.7), which ranks 54 of 270 models we track. On Arena's human-vote text board it is ranked 77 of 166. Scores close together are a tie in practice.

How does Qwen 3.8 27B compare with Gemini 3 Flash?

On quality, neither model leads clearly across the 6 measures both have. 2 measures are too close to call. Their blended prices are similar ($1.09 and $1.13 per million tokens). See the full comparison for every measure.

How current are these numbers for Qwen 3.8 27B?

The newest data on this page is dated 26 Oct 2026. Scores come from Epoch AI and Arena, prices from the open LiteLLM price list.

AI Rankings

Qwen 3.8 27B: Score, Price and Context Window

Where Qwen 3.8 27B from Alibaba ranks, what it costs and how much text it can read, in plain numbers with the sources.

Data updated 26 Oct 2026

Alibaba
Open weights
Released Aug 2026

Its score range overlaps with Gemini 3 Flash, the model just above it, so a small gap in rank is not a meaningful difference.

Best for

Running it yourself: its weights can be downloaded (read the licence terms first).

These lines are worked out from the numbers on this page, not from hands-on testing.

Key facts

Overall score: 151.7; Rank 54 of 270. Likely range 149.8 to 153.7.
Human preference: chat: 1437; Rank 77 of 166. 26,238 votes.
Human preference: coding: 1499; Rank 63 of 166. 7,134 votes.
Human preference: web pages: 1591; Rank 23 of 85. 17,186 votes.
Human preference: images: 1242; Rank 51 of 85. 7,614 votes.
Price per million tokens: $0.45 in / $3.00 out; Blended $1.09. Cached input $0.095. Median across hosting providers, checked 26 Oct 2026.
Context window: 262K tokens; About 197,000 words at once. One answer up to 33K tokens.
Weights: Open; The weights can be downloaded. Check the licence before commercial use.

Features listed by the price list: Image input, Tool use, Reasoning, Structured output.

Try itQwen Studio in our AI tools directoryQwen assistant for chat, document processing, multimodal questions, web search, and creative tasks.

How Qwen 3.8 27B stacks up

Each strip shows every model we track as faint bars. The dot is this model. Better is always to the right, even for price.

Qwen 3.8 27B compared with all models we trackOverall capability: 151.7. Better than 80% of the other 269 models. Human preference: chat: 1437. Better than 53% of the other 165 models. Human preference: coding: 1499. Better than 62% of the other 165 models. Human preference: web pages: 1591. Better than 73% of the other 84 models. Human preference: images: 1242. Better than 40% of the other 84 models. Price: $1.09 per 1M. Cheaper than 50% of the other 200 priced models. Context window: 262K tokens. Larger context window than 50% of the other 198 models.

Overall capability151.7
Better than 80% of the other 269 modelsTop 25%
Epoch Capabilities Index: one score that combines many tests.
Human preference: chat1437
Better than 53% of the other 165 modelsUpper half
How often people pick its answers over another model's on Arena.
Human preference: coding1499
Better than 62% of the other 165 modelsUpper half
How often people pick its answers to coding questions on Arena.
Human preference: web pages1591
Better than 73% of the other 84 modelsUpper half
How often people pick the web pages and small apps it builds on Arena.
Human preference: images1242
Better than 40% of the other 84 models
How often people pick its answers about pictures on Arena.
Price$1.09 per 1M
Cheaper than 50% of the other 200 priced modelsUpper half
Blended price per million tokens. Lower is better, so the cheapest models sit on the right.
Context window262K tokens
Larger context window than 50% of the other 198 modelsUpper half
How much text it can read at once. Bigger suits long documents.

Compared with the other published models that have each measure. Data updated 26 Oct 2026.

Where Qwen 3.8 27B ranks

Leaderboards it appears on, best placing first.

See all leaderboards

Test results

Each test checks one kind of skill. Place shows where it stands among the models that took that test.

ARC-AGI-2
Visual puzzles where the AI must work out a new rule from a few examples and apply it. They test picking up a new idea quickly, not recalling facts.
42.4%
Place 38 of 86

What it costs in everyday terms

List prices turned into amounts you can picture. Real bills depend on how you use it.

100 chats: $0.20; Cost of 100 chats assumes each chat sends 1,000 tokens (about 750 words) and gets 500 tokens (about 375 words) back.
A 300-page book: $0.04; Cost to read a 300-page book assumes about 100,000 input tokens and no reply. A real book varies with page size and language.
Blended price: $1.09 per 1M; Three parts input to one part output. Our own calculation for comparing models.

Similar models

Models from other makers with a similar overall score.

Learn more

Questions about Qwen 3.8 27B

How much does Qwen 3.8 27B cost?: Qwen 3.8 27B costs $0.45 per million input tokens and $3.00 per million output tokens, using the median price across hosting providers as of 26 Oct 2026. That is about $0.20 for 100 chats of about 1,000 tokens in and 500 tokens back. Confirm on the maker's site before you buy, because prices change.
What is the context window of Qwen 3.8 27B?: Qwen 3.8 27B has a context window of 262K tokens, roughly 197,000 words of text read at once. A single answer can be up to 33K tokens long.
Is Qwen 3.8 27B open source?: Qwen 3.8 27B has open weights: its trained parameters can be downloaded. That is not always the same as open source, because licence terms differ, so read the licence before you use it commercially.
How good is Qwen 3.8 27B compared with other AI models?: On the Epoch Capabilities Index, Qwen 3.8 27B scores 151.7 (likely range 149.8 to 153.7), which ranks 54 of 270 models we track. On Arena's human-vote text board it is ranked 77 of 166. Scores close together are a tie in practice.
How does Qwen 3.8 27B compare with Gemini 3 Flash?: On quality, neither model leads clearly across the 6 measures both have. 2 measures are too close to call. Their blended prices are similar ($1.09 and $1.13 per million tokens). See the full comparison for every measure.
How current are these numbers for Qwen 3.8 27B?: The newest data on this page is dated 26 Oct 2026. Scores come from Epoch AI and Arena, prices from the open LiteLLM price list.

Where this data comes from

Epoch AI: Benchmark data: Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks (CC BY 4.0). We match models across sources, rank them and show them in charts. Scores are not changed. Epoch AI does not endorse this site.
Arena (formerly LMArena): Human-preference ratings: Arena (arena.ai) leaderboard dataset, CC BY 4.0. We keep the latest published ratings, match them to models from other sources and rank them. Ratings are not changed. Arena does not endorse this site.
LiteLLM model price list: Prices and context windows: LiteLLM model price list (github.com/BerriAI/litellm, MIT licence). We use the model maker's own price entry when there is one. Prices change often, so always confirm on the maker's site before you buy.

Newest data on this page: 26 Oct 2026. Prices are list prices and change often, so confirm on the maker's site before you buy. Sponsors never change a score or an order. How we calculate this · Report a mistake