Qwen3.8 27B pricing.
prices as of · 51 providers · 59 prices
Alibaba Cloud's Qwen3.8 27B holds 1M tokens of context, reads images as well as text and has a reasoning mode. Open weights, so anyone may host it, which is why the prices below differ so much.
Qwen3.8 27B costs $0.42 in and $2.55 out per million tokens from Alibaba Cloud; the cheapest price we can explain is $0.49 at NanoGPT, checked 16 Sept 2026.
Alibaba Cloud list price
$3.82/M tokens
$0.42 in · $2.55 out
read · 8% above the market median
Cheapest price we can explain
$0.49/M tokens
$0.15 in · $0.04 out · 87% under list
NanoGPTread
What it costs for your work
Alibaba Cloud list price
$0.408a month
cheapest we can explain · NanoGPT
$0.0924a month
difference
$0.316saved
A page is about 600 tokens in and 60 out, a support reply 200 in and 300 out, an email 150 in and 250 out. Nothing else is counted: no cache discount, no batch rate, no free tier.
Every provider selling Qwen3.8 27B
Four groups, decided by the price and by what the provider publishes about it. Prices are per million tokens, and the share is that row against Alibaba Cloud's price.
The maker's list price
1 priceWhat the lab that trained the model charges on its own API.
Alibaba CloudThe price Alibaba Cloud charges for the model it trained. Every row below is measured against it.
$0.42 / $2.55in / out per M
1.00×of list price
Cheaper, and we know why
32 pricesBelow list, with a published reason: hosting the open weights, a regional price list, credits, a stated discount.
- NanoGPT
Routes to a host serving the open weights; precision not disclosed.
262k context hereNanoGPT ↗$0.15 / $0.04in / out per M
0.13×of list price
- NanoGPT
Routes to a host serving the open weights; precision not disclosed.
262k context hereNanoGPT ↗$0.15 / $0.04in / out per M
0.13×of list price
CrofServes the open weights at Q4_0, a lower precision than the maker's, so not quite the same product.
$0.09 / $0.30in / out per M
0.15×of list price
RunInfraHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context here$0.10 / $0.40in / out per M
0.18×of list price
CortecsRoutes to a host serving the open weights; precision not disclosed.
262k context hereCortecs ↗$0.10 / $0.41in / out per M
0.19×of list price
Empirio LabsRoutes to a host serving the open weights; precision not disclosed.
262k context hereEmpirio Labs ↗$0.17 / $0.50in / out per M
0.26×of list price
- Crof
Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context hereCrof ↗$0.20 / $1.50in / out per M
0.55×of list price
DeepInfraHosts the open weights on its own servers, so the price is its own, not a resale. Precision bf16.
$0.15 / $1.88in / out per M
0.61×of list price
Nous PortalResells OpenRouter's catalog at 0.80× the listed price, funded by a subscription whose credits carry a 10% bonus.
$0.15 / $1.88in / out per M
0.61×of list price
- Darkbloom
Serves the open weights at fp4, a lower precision than the maker's, so not quite the same product.
262k context herefp4 precision$0.15 / $2in / out per M
0.64×of list price
LLM TechServes the open weights at nvfp4, a lower precision than the maker's, so not quite the same product.
262k context herenvfp4 precision$0.25 / $2.09in / out per M
0.74×of list price
ParasailServes the open weights at fp8, a lower precision than the maker's, so not quite the same product.
$0.24 / $2.20in / out per M
0.76×of list price
AkashMLServes the open weights at fp8, a lower precision than the maker's, so not quite the same product.
262k context herefp8 precision$0.25 / $2.20in / out per M
0.77×of list price
- DeepInfra
Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context hereDeepInfra ↗$0.20 / $2.50in / out per M
0.81×of list price
DekaLLMHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context here$0.20 / $2.50in / out per M
0.81×of list price
AKI.IOHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context here$0.30 / $2.20in / out per M
0.81×of list price
$0.21 / $2.55in / out per M
0.83×of list price
RekaServes the open weights at fp8, a lower precision than the maker's, so not quite the same product.
262k context herefp8 precision$0.21 / $2.55in / out per M
0.83×of list price
Phala (RedPill)Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context herePhala (RedPill) ↗$0.24 / $2.50in / out per M
0.84×of list price
- Kosmik Compute
Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context here$0.35 / $2.20in / out per M
0.85×of list price
IonstreamServes the open weights at fp8, a lower precision than the maker's, so not quite the same product.
262k context herefp8 precision$0.28 / $2.55in / out per M
0.89×of list price
OrcaRouterHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
262k context here$0.33 / $2.40in / out per M
0.89×of list price
IteraComputeHosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.
328k context here$0.30 / $2.50in / out per M
0.89×of list price
ChutesHosts the open weights on its own servers, so the price is its own, not a resale. Precision tee.
262k context hereChutes ↗$0.32 / $2.50in / out per M
0.90×of list price
- Chutes
Serves the open weights at fp8, a lower precision than the maker's, so not quite the same product.
$0.32 / $2.50in / out per M
0.90×of list price
showing the 25 cheapest of 32 · every row in the open dataset
Same price as the maker
17 pricesThe same price the maker charges, sold by someone else.
$0.50 / $1.71in / out per M
0.84×of list price
$0.42 / $2.55in / out per M
1.00×of list price
Regolo AI36% above Alibaba Cloud's price.
120k context here$0.58 / $2.42in / out per M
1.09×of list price
$0.42 / $3in / out per M
1.11×of list price
$0.42 / $3in / out per M
1.11×of list price
$0.42 / $3in / out per M
1.11×of list price
$0.99 / $1.49in / out per M
1.17×of list price
Charm Hyper18% above Alibaba Cloud's price.
$0.50 / $3in / out per M
1.18×of list price
$0.50 / $3in / out per M
1.18×of list price
$0.50 / $3in / out per M
1.18×of list price
$0.46 / $3.12in / out per M
1.18×of list price
$0.45 / $3.20in / out per M
1.19×of list price
$0.45 / $3.20in / out per M
1.19×of list price
$0.44 / $3.30in / out per M
1.21×of list price
Models Think Evroc105% above Alibaba Cloud's price.
262k context here$0.87 / $3.50in / out per M
1.60×of list price
$0.80 / $4in / out per M
1.67×of list price
$1.20 / $4.50in / out per M
2.12×of list price
Cheaper, and nobody says why
6 prices · never rankedBelow list on a closed model with nothing to account for it. Shown, never ranked, never a pick.
TrustedRouter0.11× the list price, from one source and far below what hosting the weights costs. Listed until a second source agrees.
262k context hereTrustedRouter ↗$0.05 / $0.34in / out per M
0.13×of list price
MeganovaLists closed models at 0.80× the maker's price across its catalog.
66k context hereMeganova ↗$0.35 / $2.55in / out per M
0.94×of list price
Berget AIWe could not read this provider's prices reliably, so the row is listed but never ranked.
$0.40 / $3in / out per M
1.10×of list price
FeatherlessWe could not read this provider's prices reliably, so the row is listed but never ranked.
262k context hereFeatherless ↗$0.40 / $3in / out per M
1.10×of list price
MancerWe could not read this provider's prices reliably, so the row is listed but never ranked.
$2.25 / $2.50in / out per M
2.42×of list price
AmbientWe could not read this provider's prices reliably, so the row is listed but never ranked.
33k context hereAmbient ↗$320,000 / $3,200,000in / out per M
1087581.70×of list price
Given away with limits
3 free tiers · never rankedThese providers give Qwen3.8 27B away inside a quota of their own. A limit is not a price, so none of them counts as cheapest here or anywhere else on the site.
OVHcloud AI Endpoints
AMD
Hetzner
Frequently asked
- Who sells Qwen3.8 27B cheapest?
- NanoGPT, at $0.15 in and $0.04 out per million tokens, read on 16 Sept 2026. Routes to a host serving the open weights; precision not disclosed.
- Is Qwen3.8 27B open weights?
- Yes. Alibaba Cloud publishes the weights, so anyone may host Qwen3.8 27B, which is why 51 providers sell it and why their prices differ so much.
- How much context does Qwen3.8 27B hold?
- 1,000,000 tokens in one call, which is about 1M, and it can call your own functions.
- How does Qwen3.8 27B compare with the rest of the market?
- At $3.82 per million tokens, Alibaba Cloud's price is 8% above the $3.55 median across the 337 text models we track.