Private AI

Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Nano GPT logo
NanoGPT
Search
 
 
 
 
All conversations

Create

Chat
Studio

Explore

Models
Pricing
AI Apps
Blog

Library

Gallery
Workspace
Usage & billing
API
Suggestions and bugs
Use cases

|

Terms

|

Privacy

|

Refunds
Usage & billing
Teams
Referrals
Presets
Settings
FAQPress & mediaBrand assetsBenchmarksNano RPCEarn crypto
  1. Models
  2. Text
  3. Qwen

Qwen3.8 27B Thinking

Qwen3.8 27B is an open-weight multimodal model from Alibaba for reasoning, coding, visual understanding, tool use, and structured output. This variant enables thinking by default. Requests containing video cost $0.32 per million input tokens and $2.50 per million output tokens.

  • Reasoning
  • Vision
  • Video Input
  • Tool Calling
  • Structured Output

Added Aug 16, 2026

Model weights

Pricing

Auto routing · per 1M tokens
Input
$0.15
Output
$0.70
Cache read
$0.04
Compare provider prices

Specifications

Context window
1M
Max output
65.5K
Parameters
27B
Avg output (7d)
1.5K tokens
Longer than 84% of models

Benchmarks

No public benchmark scores for this model yet.

Providers

Choose or restrict providers with the API. Provider-selection prices include the selection markup. Provider selection guide · Pricing API (JSON, USD per 1K tokens)

Cache write /M
Auto
Auto (ZDR)
$0.20$1.40$0.04N/AN/AN/A
Auto
Auto (Anonymized)
$0.15$0.70$0.04N/AN/AN/A
Creekside Compute
Creekside Compute
NVFP4
GLB
$0.0229$0.309$0.0126N/AN/AN/A
Munito
Munito
89% off
NVFP4
EU
$0.21$0.0231$2.84$0.312$0.116$0.0127N/AN/AN/A
Darkbloom
Darkbloom
TEE attested
INT4
GLB
$0.0367$1.57N/AN/A0.6s44 tps
Reka
Reka
UNK
$0.0367$0.331$0.0189N/AN/AN/A
Chutes
Chutes
FP8
US
$0.041$1.31$0.0041N/A3.1s36 tps
DekaLLM
DekaLLM
ID
$0.042$2.63$0.021N/A0.8s99 tps
Modal
Modal
US
$0.042$2.42$0.021N/A0.5s125 tps
EngyAI
EngyAI
GLB
$0.0473$0.336$0.0158N/AN/AN/A
Wafer
Wafer
FP8
US
$0.063$2.42$0.0525N/A0.7s136 tps
PrimeIntellect
PrimeIntellect
FP8
UNK
$0.0872$0.618$0.00945N/A0.6s36 tps
Entrim
Entrim
FP8
EU
$0.0998$0.399$0.0399N/AN/AN/A
Aoru AI
Aoru AI
NVFP4
KR
$0.105$0.63$0.0525N/AN/AN/A
Astorias
Astorias
FP8
IN
$0.105$1.68$0.0525N/AN/AN/A
Hyperfusion
Hyperfusion
BF16
AE
$0.105$1.05N/AN/AN/AN/A
Uomi
Uomi
BF16
GLB
$0.126$1.57N/AN/AN/AN/A
DeepInfra
DeepInfra
BF16
US
$0.157$1.97$0.0394N/A0.8s32 tps
Mancer
Mancer
FP8
$0.157$1.31$0.0787N/A1.1s72 tps
StreamLake
StreamLake
65% off
CN
$0.525$0.184$3.15$1.10$0.105$0.0368N/A25s17 tps
Akash
Akash
FP8
GLB
$0.236$2.08$0.0525N/A0.9s47 tps
Parasail
Parasail
FP8
US
$0.252$2.31$0.0525N/A0.6s51 tps
Mixlayer
Mixlayer
BF16
GLB
$0.263$2.05$0.084N/AN/AN/A
CoreWeave
CoreWeave
FP8
US
$0.42$3.15$0.157N/A1s22 tps
Novita
Novita
US
$0.441$3.15$0.0892N/A3.3s42 tps
Alibaba
Alibaba
SG
$0.446$2.68$0.0892$0.5582.2s52.5 tps
Cloudflare
Cloudflare
US
$0.472$3.36$0.0525N/A0.4s41.5 tps
GMICloud
GMICloud
BF16
US
$0.472$3.36$0.263N/A2s41 tps
IONOS
IONOS
EU
$0.472$2.84N/AN/AN/AN/A
Nebius
Nebius
FP4
GLB
$0.472$3.15$0.105N/A0.9s175 tps
NeuralWatt
NeuralWatt
FP8
US
$0.472$3.36$0.263N/AN/AN/A
Venice
Venice
FP8
US
$0.472$3.36N/AN/A0.7s68.5 tps
Cerebras
Cerebras
FP16
US
$1.04$1.56N/AN/A0.3s821 tps
Heabsy
Heabsy
NVFP4
EU
$1.05$3.15$0.0799N/A0.9s176 tps
Privacy:ZDR, prompts not retainedNo training on promptsRetention and training unclearBest latency and throughput are highlighted.
Batch API
ProviderInputOutputCached input
Aoru AI
Aoru AIqwen3.8-27b:thinking
$0.05/M$0.30/M$0.025/M

Related text models

Compare Qwen3.8 27B Thinking with similar models from the same provider or model family.

Qwen 3.8 27B Antislop

qwen/qwen3.8-27b-antislop

Qwen 3.8 27B Antislop is a creative-writing finetune designed to reduce formulaic prose. Available capabilities include image understanding, tool calling, and optional reasoning, with context windows up to 262,144 tokens.

Qwen 3.8 27B DarkIdol

qwen/qwen3.8-27b-darkidol

Qwen 3.8 27B DarkIdol is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

Qwen 3.8 27B Fimi

qwen/qwen3.8-27b-fimi

Qwen 3.8 27B Fimi is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

Qwen 3.8 27B Kimiko

qwen/qwen3.8-27b-kimiko

Qwen 3.8 27B Kimiko is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

Qwen 3.8 27B Kiwen

qwen/qwen3.8-27b-kiwen

Qwen 3.8 27B Kiwen is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.

Qwen 3.8 27B Novelist

qwen/qwen3.8-27b-novelist

Qwen 3.8 27B Novelist is a creative finetune served by Aoru in NVFP4 with a 262,144-token context window. It supports image understanding, tool calling, and optional reasoning.