BroadRouter

Qwen3.8 27B API pricing

Updated

Among the 15 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest blended standard price for Qwen3.8 27B is Chutes at USD 0.039 per 1M input tokens and USD 1.25 per 1M output tokens. The median across 15 standard prices is USD 0.40 / USD 3. Alibaba Cloud's own API charges USD 0.50 / USD 3.

Qwen · slug qwen3.8-27b · up to 1,000,000 tokens of context · 15 providers with a standard price. Every price is read from the provider's own pricing page and linked in its row, checked 2026-11-15. No aggregator feed. Download all prices (every provider and model, with sources, CSV or JSON). How it compares with the market: the LLM API price index.

What does Alibaba Cloud charge for Qwen3.8 27B?

Input
USD 0.50
Cached input
USD 0.10
Output
USD 3
Blended 3:1
USD 1.125

USD per 1M tokens on Alibaba Cloud Model Studio, model qwen3.8-27b, standard tier. Source, checked 2026-11-14.

Is a host cheaper than Alibaba Cloud? Yes. Chutes charges USD 0.039 input and USD 1.25 output per 1M tokens, 70% below Alibaba Cloud's own USD 0.50 / USD 3 on the 3:1 blend.

Qwen3.8 27B model facts

Facts from Qwen's model card for Qwen/Qwen3.8-27B and the Hugging Face model API, read 2026-11-16. BroadNet did not verify them, and hosts may serve the model at a different precision. BroadRouter is BroadNet's site.

Qwen3.8 27B: facts as stated by the model card and the Hugging Face model API, with sources
FactAs statedSource
Open weightsYes. The weights are in a public Hugging Face repository (Qwen/Qwen3.8-27B) with no access gate.Hugging Face model API, checked 2026-11-16
Quote
“"private": false, "gated": false, 18 .safetensors files listed”
LicenceApache 2.0 (apache-2.0)Qwen model card metadata, checked 2026-11-16
Quote
“license: apache-2.0”
Total parameters27BQwen model card, checked 2026-11-16
Quote
“Number of Parameters: 27B”
Context length262,144 tokens natively, extensible up to 1,000,000Qwen model card, checked 2026-11-16
Quote
“Context Length: 262,144 natively and extensible up to 1,000,000 tokens.”
Input and outputText, images and video in; text outQwen model card, checked 2026-11-16
Quote
“a native vision-language model that understands images and videos”
Tool callingNot found on Qwen's model card when we read itchecked 2026-11-16
First published on Hugging Face (not necessarily the release date)2026-08-05Hugging Face model API, checked 2026-11-16
Quote
“"createdAt": "2026-08-05T08:22:59.000Z"”

Qwen3.8 27B price per provider

USD per 1M tokens. Standard prices are ranked on a blend of 3 input tokens to 1 output token, cheapest first; the dearest standard price is 4.7× the cheapest. Free rows, the maker's serving version and other service tiers follow, tagged and unranked.

Qwen3.8 27B API prices by provider
#ProviderInputCached inputOutputBlended 3:1ContextPrecisionSource
1Chutes cheapest
Qwen/Qwen3.8-27B-TEE
Notes and source quote

Runs on confidential TEE compute.

“Qwen3.8-27B Qwen/Qwen3.8-27B-TEE $0.0039 per 1M tokens $0.039 per 1M tokens $1.25 per 1M tokens 262K”
USD 0.039USD 0.0039USD 1.25USD 0.3417262,144FP8Chutes page
checked 2026-11-15
re-checked 2026-11-16
2DeepInfra promo
Qwen/Qwen3.8-27B
Notes and source quote

Promotional price, 25% off list (list 0.2 in / 2.5 out).

“$0.15 in $1.875 out $0.038 cached / 1M tokens”
USD 0.15USD 0.0375USD 1.875USD 0.5813262,144not statedDeepInfra page
checked 2026-11-14
re-checked 2026-11-16
3Hyperbolic
qwen3.8-27b
Notes and source quote

Context shown on page as "196K" (rounded); stored as the decimal value.

“Qwen3.8 27B qwen3.8-27b Alibaba · 196K context · Text $0.20 $2.00 $0.02”
USD 0.20USD 0.02USD 2USD 0.65196,000not statedHyperbolic page
checked 2026-11-14
re-checked 2026-11-16
4AkashML promo
Qwen/Qwen3.8-27B
Notes and source quote

Labelled "Limited Trial" on AkashML.

“Price (per 1M Tokens) Input Output Cache Read QWEN3.8 27B $0.225 $1.98 $0.05”
USD 0.225USD 0.05USD 1.98USD 0.6638262,144not statedAkashML page
checked 2026-11-15
re-checked 2026-11-16
5Crusoe Cloud
qwen/Qwen3.8-27B
Notes and source quote
“Qwen3.8 27B text, image $0.25 $2.00 $0.15”
USD 0.25USD 0.15USD 2USD 0.6875262,144not statedCrusoe Cloud page
checked 2026-11-15
re-checked 2026-11-16
6Nscale
Notes and source quote
“| Qwen3.8-27B | Qwen | $0.20 | $2.50 |”
USD 0.20not publishedUSD 2.50USD 0.775not statednot statedNscale page
checked 2026-11-15
re-checked 2026-11-16
7CoreWeave Forge
Qwen/Qwen3.8-27B
Notes and source quote
“Qwen3.8 27B $0.40 $3.00 $0.15”
USD 0.40USD 0.15USD 3USD 1.05262,144not statedCoreWeave Forge page
checked 2026-11-15
re-checked 2026-11-16
8Featherless
Qwen/Qwen3.8-27B
Notes and source quote

Developer plan (per-token billing, from 50 USD a month in credits).

“PRICING: INPUT $0.4 / CACHED $0.15 / OUTPUT $3”
USD 0.40USD 0.15USD 3USD 1.05262,144FP8Featherless page
checked 2026-11-15
re-checked 2026-11-16
9Novita AI
qwen/qwen3.8-27b
Notes and source quote
“Qwen3.8 27B 977K $0.42 /Mt · Cache Read $0.085 /Mt $3 /Mt”
USD 0.42USD 0.085USD 3USD 1.0651,000,000not statedNovita AI page
checked 2026-11-14
re-checked 2026-11-16
10Cerebras
qwen-3.8-27b
Notes and source quote

Developer (pay-as-you-go) tier. Cached input billed at the standard input rate. Paid context 131,072 tokens (inference-docs.cerebras.ai/models/qwen-3.8-27b).

“Qwen 3.8 27B ~1,850 tokens/s $0.99/M tokens $1.49/M tokens”
USD 0.99USD 0.99USD 1.49USD 1.115131,072not statedCerebras page
checked 2026-11-14
re-checked 2026-11-16
11Alibaba Cloud maker
qwen3.8-27b
Notes and source quote

International (Singapore) deployment. Open-weight model. Prompts up to 1M tokens. Cached price derived from the Context Cache page (https://www.alibabacloud.com/help/en/model-studio/context-cache), checked 2026-11-15: implicit-cache hits bill at 20% of the input price in the International (Singapore) scope.

“qwen3.8-27b context caching discount International Non-Thinking and Thinking modes 0<Token≤1M $0.5 $3”
USD 0.50USD 0.10USD 3USD 1.125not statednot statedAlibaba Cloud page
checked 2026-11-14
re-checked 2026-11-16
12Cloudflare Workers AI
@cf/qwen/qwen3.8-27b
Notes and source quote

Cloudflare publishes this USD per-token price alongside the neuron price.

“@cf/qwen/qwen3.8-27b $0.450 per M input tokens $3.200 per M output tokens”
USD 0.45not publishedUSD 3.20USD 1.138not statednot statedCloudflare Workers AI page
checked 2026-11-14
re-checked 2026-11-16
13GMI Cloud
Qwen/Qwen3.8-27B
Notes and source quote
“Price (per 1M tokens) Input $0.45 Output $3.20 Cache Read $0.25 Cache Write -”
USD 0.45USD 0.25USD 3.20USD 1.138262,144FP8GMI Cloud page
checked 2026-11-15
re-checked 2026-11-16
14Venice
qwen-3-8-27b
Notes and source quote

Beta model.

“| Qwen 3.8 27B (Beta) | `qwen-3-8-27b` | $0.45 | $3.20 | — | — | Private |”
USD 0.45not publishedUSD 3.20USD 1.138262,144FP8Venice page
checked 2026-11-15
re-checked 2026-11-16
15Groq
qwen/qwen3.8-27b
Notes and source quote

Preview model (evaluation use only per Groq).

“qwen/qwen3.8-27b | 450 | $0.80 input$4.00 output | 250K TPM1K RPM | 131,072”
USD 0.80not publishedUSD 4USD 1.60131,072not statedGroq page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra flex / spot promo
Qwen/Qwen3.8-27B
Notes and source quote

Flex service tier (0.8× standard). Promotional price, 25% off list (list 0.2 in / 2.5 out).

“Flex (0.8×) $0.12 $1.50 $0.03”
USD 0.12USD 0.03USD 1.50USD 0.465262,144not statedDeepInfra page
checked 2026-11-14
re-checked 2026-11-16
—DeepInfra priority promo
Qwen/Qwen3.8-27B
Notes and source quote

Priority service tier (1.5× standard). Promotional price, 25% off list (list 0.2 in / 2.5 out).

“Priority (1.5×) $0.225 $2.8125 $0.0563”
USD 0.225USD 0.0563USD 2.8125USD 0.8719262,144not statedDeepInfra page
checked 2026-11-14
re-checked 2026-11-16
—BroadNet Inference this site's owner
broadnet/qwen3.8-27b
Same model. Not ranked: BroadRouter is BroadNet's own site. Model page
USD 0.175USD 0.0435USD 2.17USD 0.6737262,144not statedpricing.json
synced 2026-11-14

Report a wrong price or number (email to info@broadnet.ai, with this page's link filled in).

What does Qwen3.8 27B cost for a real workload?

Monthly cost over 30 days at each standard price, using the presets of the LLM API cost calculator. Providers without a cached price bill cached input at the input price.

  • Support chatbot: 2,000 requests a day, 1,500 input tokens (60% cached), 300 output.
  • RAG over documents: 1,000 requests a day, 8,000 input tokens (30% cached), 500 output.
  • Coding agent (one developer): 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning.
Monthly cost of three workloads on Qwen3.8 27B, by provider
ProviderSupport chatbotRAG over documentsCoding agent (one developer)
ChutesUSD 24.11USD 25.58USD 129.00
DeepInfra (promo)USD 41.18USD 56.02USD 266.63
HyperbolicUSD 44.28USD 65.04USD 264.60
AkashML (promo)USD 46.44USD 71.10USD 315.45
Crusoe CloudUSD 53.10USD 82.80USD 477.00
NscaleUSD 63.00USD 85.50USD 585.00
CoreWeave ForgeUSD 76.50USD 123.00USD 607.50
FeatherlessUSD 76.50USD 123.00USD 607.50
Novita AIUSD 73.71USD 121.68USD 513.45
CerebrasUSD 115.92USD 259.95USD 1,916
Alibaba Cloud (maker)USD 77.40USD 136.20USD 558.00
Cloudflare Workers AIUSD 98.10USD 156.00USD 1,098
GMI CloudUSD 87.30USD 141.60USD 792.00
VeniceUSD 98.10USD 156.00USD 1,098
GroqUSD 144.00USD 252.00USD 1,800
BroadNet Inference (this site's owner)USD 47.71USD 65.08USD 309.11

For the support chatbot, the cheapest price (Chutes, USD 24.11 a month) saves USD 52.39 against the median (USD 76.50). For the coding agent the gap is USD 456.00 a month (78% lower).

Which Qwen3.8 27B provider is cheapest for your traffic?

Chutes has the lowest input price (USD 0.039) and the lowest output price (USD 1.25), so it is the cheapest standard price at every mix of input and output tokens, before cache discounts.

Qwen3.8 27B: price per 1M tokens at 3 input to 1 output, by share of input served from cache
ProviderNo cache hits50% cached90% cached
ChutesUSD 0.3417 cheapestUSD 0.3286 cheapestUSD 0.3181 cheapest
DeepInfraUSD 0.5813USD 0.5391USD 0.5053
HyperbolicUSD 0.65USD 0.5825USD 0.5285
AkashMLUSD 0.6638USD 0.5981USD 0.5456
Crusoe CloudUSD 0.6875USD 0.65USD 0.62
Nscale (no cache price)USD 0.775USD 0.775USD 0.775

USD per 1M tokens, 3 input tokens to 1 output token, each provider's cheapest standard row; 9 dearer providers are in the price table above. A provider without a cached price bills cached input at the input price. Any charge for writing to the cache, which some providers add, is not included.

10 of 15 providers publish a cached-input price below the input price: Chutes (90% off), DeepInfra (75% off), Hyperbolic (90% off), AkashML (78% off), Crusoe Cloud (40% off) and 5 more. Nscale, Cloudflare Workers AI, Venice and Groq publish none, so cached tokens there bill at the input price. Cerebras lists a cached price equal to its input price.

Context windows differ: 1,000,000 tokens at Novita AI; 262,144 tokens at Chutes, DeepInfra, AkashML and 5 more; 196,000 tokens at Hyperbolic; 131,072 tokens at Cerebras and Groq. Nscale, Alibaba Cloud and Cloudflare Workers AI do not state one. The cheapest row is not always the one that fits a long prompt.

Qwen3.8 27B on BroadNet Inference

BroadRouter is BroadNet's own site. BroadNet Inference serves this model as broadnet/qwen3.8-27b at USD 0.175 input, USD 0.0435 cached input and USD 2.17 output per 1M tokens, with a 262,144-token context. On the 3:1 blend, 4 standard prices above are lower, so pick BroadNet for other reasons: zero data retention and one OpenAI-compatible key for every BroadNet model.

Qwen3.8 27B on BroadNet: setup, limits and code

What does Qwen3.8 27B cost in other languages?

The same text needs a different number of tokens in each language. With the Qwen3.8 27B tokenizer, sending 1M characters costs this much at the cheapest standard input price we found for Qwen3.8 27B:

ContentInput per 1M charactersCheapest provider
EnglishUSD 0.0087Chutes
Arabic MSAUSD 0.012Chutes
HindiUSD 0.016Chutes
ChineseUSD 0.023Chutes
CodeUSD 0.011Chutes

From BroadRouter's real cost by language dataset, which covers 15 languages and content types for every model we could match to a measured tokenizer.

Published benchmark results

BroadNet did not run these benchmarks; results are as published by the Qwen model card and may use different settings.

Published benchmark results for Qwen3.8 27B, with sources
BenchmarkResultSetting or noteSource
GPQA Diamond89.2not statedQwen model card, checked 2026-11-16
LiveCodeBench v690.3not statedQwen model card, checked 2026-11-16
SWE-bench Pro61.7not statedQwen model card, checked 2026-11-16
Terminal Bench 2.1 (Terminus)73.0not statedQwen model card, checked 2026-11-16
IFBench79.5not statedQwen model card, checked 2026-11-16
OSWorld-Verified84.3not statedQwen model card, checked 2026-11-16
OmniDocBench 1.591.1not statedQwen model card, checked 2026-11-16

Estimate your own bill

The Qwen models whose cheapest standard price is closest to Qwen3.8 27B's (USD 0.3417 blended):

Qwen models priced closest to Qwen3.8 27B
ModelCheapest inputCheapest outputMax contextProviders
Qwen3 Next 80B A3B InstructUSD 0.09USD 1.10262,1445
Qwen3.5 35B A3BUSD 0.14USD 1262,1447
Qwen2.5-72B-InstructUSD 0.36USD 0.4032,7683
Qwen3 VL 235B A22BUSD 0.20USD 0.88262,1443
Qwen-MT PlusUSD 0.25USD 0.7516,3842
Qwen3-VL 8BUSD 0.18USD 0.68262,1442

Priced closest from other makers: DeepSeek-V4-Flash-Vision-Exp (DeepSeek, USD 0.3234) · MiniMax M2.7 (MiniMax, USD 0.3675) · DeepSeek R1 Distill Llama 70B (DeepSeek, USD 0.375).

Compare Qwen3.8 27B with

Also compared: vs Qwen3.6 27B · vs DeepSeek V4.1 Flash · vs Gemma 4 31B · vs GLM 5.3 · vs OpenAI GPT OSS 120B · vs Llama 4 Maverick 17B 128E Instruct FP8 · vs MiniMax M3 · vs DeepSeek V3.2 · vs Qwen3.8 2.4T A95B · vs DeepSeek-V4-Flash-Vision-Exp · vs GLM 4.5 Air · vs Step 3.7 Flash · vs MiniMax M3.1 Flash · vs GPT-5.6 Luna · vs Inkling 2 Small.

Every Qwen3.8 27B price comparison · All model-vs-model price comparisons

All 429 models: API prices compared · Cheapest LLM APIs right now · LLM API price index (open data) · Download all prices (CSV, JSON) · Price change log

Questions

How much does the Qwen3.8 27B API cost per token?

From USD 0.039 per 1M input tokens (USD 0.000000039 per token) and USD 1.25 per 1M output tokens (USD 0.00000125 per token) at the lowest standard prices we found. The median of 15 standard prices is USD 0.40 input and USD 3 output. Each price was read from the provider's own pricing page; the source is linked in its row.

Which provider is cheapest for Qwen3.8 27B?

Among the 15 providers we checked (prices read 2026-11-14 to 2026-11-15), the lowest is Chutes, ranked on a blend of 3 input tokens to 1 output token (USD 0.3417 per 1M). It also has the lowest input and the lowest output price. Batch, flex, priority and fast tiers are shown in the table but not ranked.

Is Qwen3.8 27B cheaper from a host than from Alibaba Cloud?

Yes. Chutes charges USD 0.039 input and USD 1.25 output per 1M tokens, 70% below Alibaba Cloud's own USD 0.50 / USD 3 on the 3:1 blend.

Does Qwen3.8 27B have cached input pricing?

11 of 15 standard prices include one. The lowest is USD 0.0039 per 1M at Chutes; the median is USD 0.10. Providers without one bill cached input at the input price. What cached input means.

What is the Qwen3.8 27B context window?

Up to 1,000,000 tokens. 11 standard prices come with less, down to 131,072 tokens (Cerebras, Groq, Hyperbolic, AkashML and 7 more). Check the context column before you pick the cheapest row for long prompts. Context window calculator.

How much does a coding agent cost per month on Qwen3.8 27B?

For 1,500 requests a day, 40,000 input tokens (85% cached), 2,000 output incl. reasoning, over 30 days: USD 129.00 a month at Chutes, USD 585.00 at the median price and USD 1,916 at Cerebras. Try your own numbers.

Is there a batch, flex or priority price for Qwen3.8 27B?

Yes: flex / spot from DeepInfra at USD 0.12 / USD 1.50; priority from DeepInfra at USD 0.225 / USD 2.8125 per 1M input / output tokens. These tiers trade speed or availability for price (batch runs within hours, flex and spot can be slow or refused at capacity, priority and fast cost more for speed or reliability), so they are shown but not ranked.

Is Qwen3.8 27B free to use through an API?

No provider in our data lists a free price for Qwen3.8 27B as of 2026-11-15. BroadNet Inference serves it as broadnet/qwen3.8-27b to paid accounts only: the USD 5 starting credit does not cover it (rate card synced 2026-11-14). Free tiers and credits by provider (checked 2026-11-14) lists general offers that do not name a model; all free LLM API models are on one page.

Is Qwen3.8 27B open source, with open weights, and what is its licence?

Qwen publishes the weights on Hugging Face as Qwen/Qwen3.8-27B, in a public repository with no access gate. The licence, as the Qwen model card metadata states it: Apache 2.0 (apache-2.0). Open weights means you can download and run the model yourself; what you may do with it, including commercial use, is set by that licence, so read it in full (checked 2026-11-16). These facts come from Qwen's card and the Hugging Face model API; BroadNet did not verify them, and hosts may serve the model at a different precision.

How many parameters does Qwen3.8 27B have?

Total parameters, as the Qwen model card states them: 27B. (checked 2026-11-16; BroadNet did not verify these figures.)

Can I use Qwen3.8 27B on BroadNet Inference?

Yes. BroadNet Inference serves Qwen3.8 27B as broadnet/qwen3.8-27b at USD 0.175 input, USD 0.0435 cached and USD 2.17 output per 1M tokens (rate card synced 2026-11-14), with zero data retention. On the 3:1 blend, 4 standard prices on this page are lower. BroadRouter is BroadNet's own site. Qwen3.8 27B on BroadNet.

How do I call Qwen3.8 27B with the OpenAI SDK?

On BroadNet Inference, point the OpenAI SDK at https://inference.broadnet.ai/v1 with your BroadNet key and set the model to broadnet/qwen3.8-27b: client = OpenAI(base_url="https://inference.broadnet.ai/v1", api_key=...), then client.chat.completions.create(model="broadnet/qwen3.8-27b", messages=[...]). BroadRouter is BroadNet's own site. Setup notes for agents: llms.txt for Qwen3.8 27B on BroadNet.

How does Qwen3.8 27B score on benchmarks?

As published, the Qwen model card reports 89.2 on GPQA Diamond, 90.3 on LiveCodeBench v6 and 61.7 on SWE-bench Pro (checked 2026-11-16). BroadNet did not run these benchmarks, and sources may use different settings, so the figures are not directly comparable. Every published result we found, with its setting and source.

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity