Roboflow

Qwen3.7 Plus vs Qwen3.8 27B

Compare Qwen3.7 Plus and Qwen3.8 27B side-by-side. See how these vision models stack up in Image Captioning, Open Prompt, Classification, Object Detection, and OCR.

Compare Qwen3.7 Plus vs Qwen3.8 27B live

Run the same image across every model that supports a task and compare their outputs side-by-side.

Detect and compare bounding boxes across models on the same image.

Open Object Detection in the full playground
QwenQwen3.7 Plus
Run to compare this model.
QwenQwen3.8 27B
Run to compare this model.

Models in this comparison

Qwen3.7 Plus vs Qwen3.8 27B on Vision Evals

Qwen3.7 Plus scores higher on all six Vision Evals tasks.

The widest gap is Counting, where Qwen3.7 Plus leads 50.0% to 41.9%.

Overall, Qwen3.7 Plus averages 67.4% (#18 of 31) against 61.2% (#30 of 31) for Qwen3.8 27B.

Qwen3.7 Plus is both cheaper ($0.0008 vs $0.0016 per sample) and faster (7.0s vs 7.3s per sample).

Qwen3.7 PlusQwen3.8 27B

Qwen3.7 Plus vs Qwen3.8 27B Comparison Table

Evals updated August 20, 2026Pricing updated August 24, 2026

PropertyQwen3.7 PlusQwen3.8 27B
OrganizationQwenQwen
Categoryclosedopen
Modalitymultimodal
Release DateAug 2026
Context Window262K
Parameters27.78B
LicenseApache 2.0
Pricing per 1M tokens
Input $/1M$0.320$0.400
Output $/1M$1.28$3.00
Vision Tasks
CaptioningDemoDemo
ClassificationDemoDemo
object-detectionDemoDemo
OCRDemoDemo
Visual Question AnsweringDemoDemo
Chart Question Answering
Document Question Answering
Image Tagging
Multi-Label Classification
Vision Language
Model Features
Foundation Vision
LLMs with Vision Capabilities
Multimodal Vision
Vision Evalsground-truth scores across 6 vision tasks, pooled at low effort
Overall
67.4%
61.2%
Avg cost / sample$0.0008$0.0016
Avg speed / sample7.01s7.33s
By task
Object Detection
60.1%
$0.0013
54.5%
$0.0033
Counting
50.0%
$0.0004
41.9%
$0.0005
Identification
84.4%
$0.0003
78.1%
$0.0004
OCR
86.5%
$0.0009
81.4%
$0.0018
Data Extraction
83.5%
$0.0004
79.4%
$0.0005
Reasoning (low)
39.7%
$0.0003
31.8%
$0.0004
Reasoning (high)
68.2%
$0.0043
62.3%
$0.0081

Qwen3.7 Plus vs Qwen3.8 27B: Overview

Qwen3.7 Plus
No description available
Qwen3.8 27B

Qwen3.8-27B is a dense vision-language model of roughly 27.78 billion parameters from Alibaba's Qwen team, built on the architectural foundation established by the Qwen3.5 generation. It takes interleaved text, image, and video input through an integrated vision encoder and produces text, covering document and chart understanding, visual question answering, front-end code generation from visual references, and agentic computer-use workflows. The decoder is a hybrid stack of 64 layers that repeats a pattern of three Gated DeltaNet linear-attention blocks, each followed by a feed-forward block, then one grouped-query full-attention block, so only a quarter of the layers maintain a conventional key-value cache. Native context length is 262,144 tokens.

Post-training targets long-horizon agentic work, and the model exposes per-request thinking control that lets callers switch reasoning traces on or off and choose a reasoning effort level. Multi-token prediction weights are trained alongside the main network for speculative decoding. Qwen reports improvements over Qwen3.6-27B on agentic and multimodal evaluations including Terminal-Bench 2.1, OSWorld-Verified, and SWE-bench Multimodal, with additional results on MathVision, CharXiv, and an internal Vision2Web suite.

Frequently Asked Questions

On Roboflow's Vision Evals, Qwen3.7 Plus performed better. It scores higher on all six vision tasks and averages 67.4% (#18 of 31) against 61.2% (#30 of 31) for Qwen3.8 27B. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.

Yes. On the Vision Evals Counting benchmark, Qwen3.7 Plus leads with 50.0% against 41.9%. This is the widest gap between the two models across the benchmark's tasks.

Qwen3.7 Plus is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0008 per sample against $0.0016. Qwen3.7 Plus is priced at $0.32 per 1M input tokens and $1.28 per 1M output; Qwen3.8 27B is priced at $0.40 per 1M input tokens and $3.00 per 1M output. Actual costs depend on your image sizes, prompts, and output length.

Qwen3.7 Plus is faster. Across Roboflow's Vision Evals it averaged 7.0s per inference against 7.3s. Latency varies with image size, prompt length, and provider load, so treat these as relative rather than guaranteed figures.

Yes. The comparison demo on this page runs both models on the same image side by side for image captioning and open prompts in the free Roboflow Playground. You can try it instantly, and a free account unlocks unlimited runs.