Skip to content
Sovyron

Qwen3.8 27B

qwen1.1M context500k max outputreleased Aug 14, 2026Open weightsReasoningTool calling

As of Nov 22, 2026, the cheapest Qwen3.8 27B API price is $0.03 per 1M input tokens (CedarLLM) and $0.32 per 1M output tokens (engy). Qwen3.8 27B is offered by 70 providers in the qwen family, open weights, reasoning enabled. 3 providers list it as a free-tier offer.

Cheapest input
$0.03
Cheapest output
$0.32
Official
—
output per 1M tokens
Providers
70
offering this model
Spread
17.5×
max / min output
Free offers
3
free-tier offers

API pricing by provider

Metered per-token prices. Cheapest input and output are highlighted; † marks a context-tier surcharge. How prices are ranked. Seeing a wide range? Why the same LLM costs different amounts.

Qwen3.8 27B API prices by provider in USD per 1M tokens, updated Nov 22, 2026
engyopenqwen3.8-27b$0.045Cheapest$0.32$0.015—1.0M—
DevPass (LLM Gateway)openqwen3.8-27b$0.08$0.35$0.05—1.0M—
Cortecsopenqwen3.8-27b$0.1$0.4$0.04—1.0M71d ago
EntrimopenQwen/Qwen3.8-27B$0.1$0.4$0.04—262k—
RunInfraopenQwen/Qwen3.8-27B$0.1$0.4$0.01—262k—
Powerspringopenqwen3.8-27b$0.1044$0.4166$0.0418—1.0M—
Liquid Inferenceopenqwen3.8-27b$0.08$0.525$0.0525—262k—
EmpirioLabs AIopenqwen3-8-27b$0.17$0.5$0.08—262k—
NanoGPTopenqwen/qwen3.8-27b$0.15$0.7$0.04—1.0M—
AKI.IOopenqwen3.8-27b$0.1675$0.7257$0.0837—262k7d agoout −70.6%
CedarLLMopenqwen3.8-27bCheapest$0.03$1.15$0.03—262k—
Obscurify.aiopenqwen/qwen3.8-27b$0.1658$0.7735$0.0442—262k—
Adaptionopenqwen3.8-27b$0.039$1.25$0.0039—262k—
Vultropenqwen3.8-27b$0.15$1——262k—
NEAR AI CloudopenQwen/Qwen3.8-27B$0.04$1.35$0.018—262k—
ParalonCloudopenqwen3.8-27b$0.12$1.7——131k—
OpenRouteropenqwen/qwen3.8-27b$0.06$1.96$0.059—1.0Myesterdayout −10.9%
Vercel AI Gatewayopenalibaba/qwen3.8-27b$0.06$1.96$0.059—262kyesterdayout −10.9%
TenstorrentopenQwen/Qwen3.8-27B$0.15$2$0.05—262k—
WaferopenQwen3.8-27B$0.06$2.3$0.05—262k—
stepping stone AGopenQwen/Qwen3.8-27B$0.2975$1.785$0.0607—262k—
LLM Techopennvidia/Qwen3.8-27B-NVFP4$0.25$2.09$0.04—262k—
Munito AIopenqwen/qwen3.8-27b$0.18$2.4$0.05—262k—
Deep InfraopenQwen/Qwen3.8-27B$0.2$2.5$0.05—262k68d ago
Phalaopenqwen/qwen3.8-27b$0.2$2.5$0.05—1.0M—
SOKKAN Routeropenqwen/qwen3.8-27b$0.2$2.5$0.05—131k—
SiliconFlow (China)openQwen/Qwen3.8-27B$0.4498$1.7992——262k—
Ofoxopenqwen/qwen3.8-27b$0.5$1.71$0.043$0.631.1M71d ago
Kosmik Computeopenqwen/qwen3.8-27b$0.35$2.2$0.09—262k—
Munito AI (EU)openqwen/qwen3.8-27b$0.2$2.7$0.11—262k—
OrcaRouteropenqwen/qwen3.8-27b$0.33$2.4——262k—
Lyceumopenqwen/qwen3.8-27b$0.4$2.4$0.1—262k—
TensorXopenqwen/qwen3.8-27b$0.4$2.4$0.1—262k—
Fortytwoopenqwen/qwen3.8-27b$0.425$2.55$0.085—262k—
Standard Computeopenqwen/qwen3.8-27b$0.425$2.55$0.085$0.5313262k—
Ambientopenqwen/qwen3.8-27b$0.32$3.2$0.16free33k—
Regolo AIopenqwen3.8-27b$0.58$2.42——120k—
ai&openqwen/qwen3.8-27b$0.4$3$0.2—262k—
CommandCodeopenQwen/Qwen3.8-27B$0.4$3$0.04—262k—
Hugging FaceopenQwen/Qwen3.8-27B$0.4$3——262k—
Merge Gatewayopenqwen/qwen3.8-27b$0.4$3$0.05—262k—
CoreWeaveopenQwen/Qwen3.8-27B$0.4$3$0.15—262k—
LLM Gatewayopennovita/qwen3.8-27b$0.42$3$0.085—1.0M—
Novita AIopenqwen/qwen3.8-27b$0.42$3$0.085—1.0M—
DoublewordopenQwen/Qwen3.8-27B-FP8$0.45$3$0.045$0.5625262k—
Nebius Token FactoryopenQwen/Qwen3.8-27B$0.45$3$0.1—262k—
Cerebrasopenqwen-3.8-27b$0.99$1.49$0.99—131k—
Eden AIopenqwen/qwen3.8-27b$0.5$3$0.1$0.6251.0M—
Charm Hyperopenqwen3.8-27b$0.5$3$0.1—1.0M—
Kilo Gatewayopenqwen/qwen3.8-27b$0.5$3$0.1$0.625262k4d agoout +17.6%
LLM Gatewayopenalibaba/qwen3.8-27b$0.5$3$0.1$0.625984k—
Cloudflare Workers AIopen@cf/qwen/qwen3.8-27b$0.45$3.2$0.05—262k—
GraphNopenqwen3.8-27b$0.45$3.2$0.045—262k—
Neuralwattopenqwen-3.8-27b$0.45$3.2$0.25—262k—
Ofoxopenbailian/qwen3.8-27b$0.45$3.2$0.05$0.56251.1M—
Prime Inferenceopenqwen/qwen3.8-27b$0.45$3.2$0.05—262k8d agoout +0%
Klavo AIopenqwen/qwen3.8-27b$0.5088$3.0285$0.0969—1.0M—
OVHcloud AI Endpointsopenqwen3.8-27b$0.47$3.19——262k65d ago
Kelnopenqwen/qwen3.8-27b$0.5$3.2$0.15—1.0M—
Opperopenqwen3.8-27b$0.5811$3——262k—
AI/ML APIopenalibaba/qwen3.8-27b$0.4761$3.3856——262k30d agoout −23.1%
Berget.AIopenQwen/Qwen3.8-27B-FP8$0.46$3.48——262k—
Syntheticopenhf:Qwen/Qwen3.8-27B$0.45$3.6$0.45—262k—
Scalewayopenqwen3.8-27b$0.684$3.762$0.137—262k—
evrocopenQwen/Qwen3.8-27B$0.87$3.5——262k—
Groqopenqwen/qwen3.8-27b$0.8$4——131k—
Thinking MachinesopenQwen/Qwen3.8-27B$0.6975$5.595$0.1395—66k—
Thinking MachinesopenQwen/Qwen3.8-27B:peft:262144$0.6975$5.595$0.1395—262k—
AMDfreeQwen3.8-27Bfreefreefree—262k—
HetznerfreeQwen3.8-27Bfreefreefree—262k—
IteraComputefreeqwen/qwen3.8-27bfreefreefree—131k—

Qwen3.8 27B price history

Output price per 1M tokens since tracking began on Sep 10, 2026, for the 6 cheapest of 10 providers whose rate moved. Each step is a recorded change. All recent price changes.

Qwen3.8 27B output price per 1M tokens by provider, Sep 10, 2026 to Nov 22, 2026$0$1.25$2.5$3.75$5Sep 10, 2026Oct 17, 2026Nov 22, 2026
Cortecs$0.4AKI.IO$0.7257OpenRouter$1.96Vercel AI Gateway$1.96Deep Infra$2.5Ofox$1.71
Qwen3.8 27B price changes by provider, newest first, USD per 1M tokens
DateProvider1M input1M outputChange
OpenRouter$0.06$2.2 → $1.96−10.9%
Vercel AI Gateway$0.06$2.2 → $1.96−10.9%
Kilo Gateway$0.425 → $0.5$2.55 → $3+17.6%
OpenRouter$0.15 → $0.06$1.875 → $2.2+17.3%
AKI.IO$0.3365 → $0.1675$2.4673 → $0.7257−70.6%
Prime Inference$0.99 → $0.45$3.2−54.5%
Vercel AI Gateway$0.06$2.3 → $2.2−4.3%
Vercel AI Gateway$0.5 → $0.06$3 → $2.3−23.3%
OpenRouter$0.039 → $0.15$1.25 → $1.875+50%
OpenRouter$0.2 → $0.039$2.5 → $1.25−50%
OpenRouter$0.04 → $0.2$1.35 → $2.5+85.2%
AI/ML API$0.6189 → $0.4761$4.4013 → $3.3856−23.1%
AKI.IO$0.3 → $0.3365$2.2 → $2.4673+12.1%
OpenRouter$0.24 → $0.04$2.2 → $1.35−38.6%
OpenRouter$0.04 → $0.24$2.3 → $2.2−4.3%
OpenRouter$0.425 → $0.04$2.55 → $2.3−9.8%
OpenRouter$0.42 → $0.425$3 → $2.55−15%
OpenRouter$0.0449 → $0.42$4.4 → $3−31.8%
OpenRouter$0.42 → $0.0449$3 → $4.4+46.7%
OpenRouter$0.2 → $0.42$2.5 → $3+20%

Qwen3.8 27B pricing FAQ

What is the cheapest Qwen3.8 27B API?

As of Nov 22, 2026, engy has the lowest Qwen3.8 27B output price at $0.32 per 1M tokens, and CedarLLM has the lowest input price at $0.03 per 1M tokens. Prices are metered per-token rates in USD; free tiers and subscription plans are excluded.

How many providers offer Qwen3.8 27B?

70 providers list Qwen3.8 27B on Sovyron; 68 of them sell it at a metered per-token price.

Is Qwen3.8 27B free?

3 providers list a free-tier offer: AMD, Hetzner, IteraCompute. Free tiers usually have rate limits.

What is the context window of Qwen3.8 27B?

Qwen3.8 27B supports a 1.1M-token context window and up to 500k output tokens.

Token cost calculator

What a month costs at one provider's published rates. Enter the tokens you expect to send and receive; cached tokens use the cached-input price where the provider lists one, and count as ordinary input where it does not. See how much prompt caching saves and why output tokens cost more than input tokens.

Estimated

—