Pricingby GlyphQwen3.8 27B is an open-weight dense vision-language model from Qwen.
Ways to buy 23
| Way to buy | Input | Output | In / out per 1M | Cached | Vs list |
|---|---|---|---|---|---|
| PrimeIntellect fp8, cheapest · cached $0.004 | $0.005 | $0.261 | $0.005 / $0.261-93% vs list | $0.004 | -93% vs list |
| Near AI fp8 · cached $0.018 | $0.04 | $1.35 | $0.04 / $1.35-62% vs list | $0.018 | -62% vs list |
| Darkbloom fp4 · cached $0.018 | $0.035 | $1.5 | $0.035 / $1.5-58% vs list | $0.018 | -58% vs list |
Default region · cached $0.02 | $0.04 | $2.3 | $0.04 / $2.3-37% vs list | $0.02 | -37% vs list |
| DekaLLM Default region · cached $0.02 | $0.04 | $2.5 | $0.04 / $2.5-32% vs list | $0.02 | -32% vs list |
| AkashML fp8 · cached $0.05 | $0.225 | $1.98 | $0.225 / $1.98-31% vs list | $0.05 | -31% vs list |
fp8 · cached $0.024 | $0.24 | $2.2 | $0.24 / $2.2-24% vs list | $0.024 | -24% vs list |
fp8 · cached $0.05 | $0.24 | $2.2 | $0.24 / $2.2-24% vs list | $0.05 | -24% vs list |
| Reka fp8 · cached $0.05 | $0.15 | $2.5 | $0.15 / $2.5-23% vs list | $0.05 | -23% vs list |
| Mancer 2 fp8 · cached $0.175 | $0.175 | $2.5 | $0.175 / $2.5-21% vs list | $0.175 | -21% vs list |
bf16 · cached $0.05 | $0.2 | $2.5 | $0.2 / $2.5-19% vs list | $0.05 | -19% vs list |
| ModelRun fp4 · cached $0.09 | $0.2 | $2.5 | $0.2 / $2.5-19% vs list | $0.09 | -19% vs list |
| Phala Default region · cached $0.05 | $0.2 | $2.5 | $0.2 / $2.5-19% vs list | $0.05 | -19% vs list |
| Ionstream fp8 · cached $0.1 | $0.25 | $2.55 | $0.25 / $2.55-14% vs list | $0.1 | -14% vs list |
The lab's own API · cached $0.085 | $0.425 | $2.55 | $0.425 / $2.55List price | $0.085 | List price |
fp8 · cached $0.15 | $0.4 | $3 | $0.4 / $3+10% vs list | $0.15 | +10% vs list |
Default region · cached $0.085 | $0.42 | $3 | $0.42 / $3+11% vs list | $0.085 | +11% vs list |
fp4 · cached $0.1 | $0.45 | $3 | $0.45 / $3+14% vs list | $0.1 | +14% vs list |
fp16 · cached $0.99 | $0.99 | $1.49 | $0.99 / $1.49+17% vs list | $0.99 | +17% vs list |
| StreamLake Default region · cached $0.1 | $0.5 | $3 | $0.5 / $3+18% vs list | $0.1 | +18% vs list |
Default region · cached $0.05 | $0.45 | $3.2 | $0.45 / $3.2+19% vs list | $0.05 | +19% vs list |
fp8 | $0.45 | $3.2 | $0.45 / $3.2+19% vs list | n/a | +19% vs list |
| Wafer fp8 · cached $0.199 | $0.2 | $6 | $0.2 / $6+73% vs list | $0.199 | +73% vs list |
USD per 1M tokens; batch $0.16 in, $2 out on the lab's API. Regions are separate routes.
Through a gateway
Your own numbers100M input and 20M output tokens a month: $93.50 direct.
Kong AI Gateway$93.50+0%
LiteLLM$93.50+0%
Vercel AI Gateway$93.50+0%
Requesty$98.17+5%
Eden AI$98.64+5.5%
OpenRouter$98.64+5.5%
Cloudflare AI GatewayNot offeredNot offered
HeliconeNot offeredNot offered
Neon AI GatewayNot offeredNot offered
TrueFoundry AI GatewayNot offeredNot offered
A dash: not on that gateway's own model list. No published token fee: Martian Gateway, Portkey.
Related models
| In / out per 1M | Cheapest route | |||||||
|---|---|---|---|---|---|---|---|---|
Qwen (Alibaba) · 1M | $4 | $12 | $4 / $12 | $6 | 1M | Lab list price | · | Qwen3.8 Max Prime is a higher-throughput variant of Qwen3.8 Max from Alibaba's Qwen team, served as a separate SKU at a higher price point. |
Qwen (Alibaba) · 1M | $0.15 | $0.47 | $0.15 / $0.47 | $0.23 | 1M | Lab list price | · | Qwen3.8 Omni Flash is an omni-modal reasoning model from Alibaba, the first Qwen model built around agentic capabilities with native audio-video understanding. |
Qwen (Alibaba) · 1M | $2 | $6 | $2 / $6 | $3 | 1M | Lab list price | · | Qwen3.8 Max 0902 is an updated snapshot of Qwen3.8 Max from Alibaba's Qwen team. |
Qwen (Alibaba) · 1M | $0.15 | $0.47 | $0.15 / $0.47 | $0.23 | 1M | Lab list price | · | Qwen3.8 Flash is a multimodal reasoning model from Alibaba. |
OpenAI · 1.05MCheapest: $1 on OpenAI (flex) | $1 | $5 | $1 / $5 | $2 | 1.05M | $1 on OpenAI (flex) | · | GPT-6.1 Terra is the mid-tier model in OpenAI's GPT-6.1 series, priced below GPT-6.1 Sol. |
Google · 1MCheapest: $0.75 on Google Gemini API (flex) | $0.75 | $3.75 | $0.75 / $3.75 | $1.5 | 1M | $0.75 on Google Gemini API (flex) | · | Gemini 3.8 Flash is Google's most intelligent Flash model with significant gains from 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning. |
Meta · 1M | $1.25 | $4.25 | $1.25 / $4.25 | $2 | 1M | Lab list price | · | Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. |
Google · 1MCheapest: $0.75 on Google Vertex AI (global-flex) | $0.75 | $3.75 | $0.75 / $3.75 | $1.5 | 1M | $0.75 on Google Vertex AI (global-flex) | · | Gemini 3.7 Flash is a multimodal model from Google for fast agentic workflows, coding, and complex multi-step reasoning. |
About Qwen3.8 27B
Takes text, image, video in and returns text. First listed Sep 23, 2026. Up to 236K output tokens. Blended price $0.956 per 1M tokens at 3 input to 1 output. Model id qwen3-8-27b.
Price history
List price since Sep 23, 2026, from daily reads of the OpenRouter catalog.
Qwen3.8 27B
Mancer 2 (fp8) set its first price: cached input $0.175 per 1M tokens.
Reka (fp8) raised its price: input $0.04 to $0.15 and output $2.2 to $2.5 per 1M tokens.
DekaLLM cut its price: cached input $0.087 to $0.02 per 1M tokens.
PrimeIntellect (fp8) cut its price: output $0.305 to $0.261 per 1M tokens.
Mancer 2 (fp8) cut its price: input $0.18 to $0.175 per 1M tokens.
PrimeIntellect (fp8) raised its price: output $0.286 to $0.305 per 1M tokens.
Darkbloom (fp4) cut its price: output $2.2 to $1.5 per 1M tokens.
DekaLLM cut its price: input $0.096 to $0.04 and output $4.4 to $2.5 per 1M tokens.
Mancer 2 (fp8) cut its price: input $0.2 to $0.18 per 1M tokens.
Across GlyphCompany profile
Questions
How much does Qwen3.8 27B cost?
$0.425 per million input tokens and $2.55 per million output tokens, with cached input at $0.085, or $0.16 and $2 through the Batch API.
What is the context window of Qwen3.8 27B?
1.02M tokens, with up to 236K tokens of output.
Where is Qwen3.8 27B cheapest?
Of 23 routes tracked, PrimeIntellect is cheapest at $0.005 input and $0.261 output per million tokens.
Has Qwen3.8 27B's price changed?
Yes. The output price went from $2.2 to $2.5, first seen Nov 3, 2026.