Roboflow

GPT-5.6 Terra vs Qwen3.5 27B

Compare GPT-5.6 Terra and Qwen3.5 27B side-by-side. See how these vision models stack up in Classification, Open Prompt, Object Detection, OCR, and Image Captioning.

Compare GPT-5.6 Terra vs Qwen3.5 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
OpenAIGPT-5.6 Terra
Run to compare this model.
QwenQwen3.5 27B
Run to compare this model.

Models in this comparison

GPT-5.6 Terra vs Qwen3.5 27B on Vision Evals

GPT-5.6 Terra scores higher on 5 of the six Vision Evals tasks.

The widest gap is Reasoning, where GPT-5.6 Terra leads 59.6% to 31.8%.

Overall, GPT-5.6 Terra averages 72.4% (#12 of 25) against 64.3% (#22 of 25) for Qwen3.5 27B.

Qwen3.5 27B is cheaper ($0.0007 vs $0.0044 per sample), while GPT-5.6 Terra is faster (7.2s vs 7.4s per sample).

GPT-5.6 TerraQwen3.5 27B

GPT-5.6 Terra vs Qwen3.5 27B Comparison Table

Evals updated August 6, 2026Pricing updated August 11, 2026

PropertyGPT-5.6 TerraQwen3.5 27B
OrganizationOpenAIQwen
Categoryclosedopen
Modalitymultimodalmultimodal
Release DateJul 2026Feb 2026
Context Window1.1M262K
Parameters27B
LicenseProprietaryApache 2.0
Pricing per 1M tokens
Input $/1M$1.00$0.195
Output $/1M$6.00$1.56
Vision Tasks
CaptioningDemoDemo
Chart Question Answering
ClassificationDemoDemo
Document Question Answering
Image Tagging
Multi-Label Classification
Object DetectionDemoDemo
OCRDemoDemo
Vision Language
Visual Question AnsweringDemoDemo
Model Features
Foundation Vision
LLMs with Vision Capabilities
Multimodal Vision
Vision Evalsground-truth scores across 6 vision tasks, pooled at low effort
Overall
72.4%
64.3%
Avg cost / sample$0.0044$0.0007
Avg speed / sample7.15s7.38s
By task
Object Detection
60.7%
$0.0070
58.8%
$0.0013
Counting
67.6%
$0.0030
54.0%
$0.0002
Identification
78.1%
$0.0020
78.1%
$0.0002
OCR
88.8%
$0.0065
84.5%
$0.0009
Data Extraction
79.4%
$0.0018
78.3%
$0.0002
Reasoning (low)
59.6%
$0.0025
31.8%
$0.0002
Reasoning (high)
64.2%
$0.0033
61.6%
$0.0065

GPT-5.6 Terra vs Qwen3.5 27B: Overview

GPT-5.6 Terra

GPT-5.6 Terra is the mid-tier reasoning model in OpenAI's GPT-5.6 family, which also includes the flagship Sol and the lightweight Luna. Introduced in a limited preview on June 26, 2026, and made broadly available on July 9, 2026, Terra accepts text and image input and produces text output, supporting vision, function calling, tool use, and agentic workflows. It is designed as a balanced option for everyday professional and production workloads — including coding assistance, document analysis, customer support, and multi-step agent tasks — where both output quality and cost efficiency matter. OpenAI positions Terra as delivering performance competitive with GPT-5.5 at approximately half the price, with a context window of around 1,050,000 tokens. On Terminal-Bench 2.1, Terra scores 84.3%, matching Claude Fable 5 on that benchmark. Under OpenAI's Preparedness Framework, Terra is rated High for cybersecurity and biological capabilities, meaning it demonstrates meaningful capability in those domains without reaching the Critical threshold.

GPT-5.6 introduces a new naming convention in which the generation number (5.6) is paired with a durable capability tier name (Sol, Terra, or Luna), allowing each tier to advance on its own schedule. Terra carries the API identifier gpt-5.6-terra and supports the same reasoning effort controls available across the family, including adjustable reasoning depth. The model includes prompt caching with explicit cache breakpoints and a 30-minute minimum cache life, with cache writes billed at 1.25x the uncached input rate and cache reads receiving a 90% discount. GPT-5.6 Terra is a proprietary, closed-weights model served through the OpenAI API, Codex, and ChatGPT.

Qwen3.5 27B

Qwen3.5-27B is a multimodal dense hybrid model developed by Alibaba Cloud’s Qwen team and released in February 2026 as a high-precision entry in the Qwen3.5 "Medium" series. Unlike its Mixture-of-Experts (MoE) siblings, the 27B model utilizes a dense architecture combining Gated Delta Networks with a feed-forward structure, activating its full parameter suite for every inference to maximize reliability. This design provides the highest instruction-following and coding accuracy in its class, with a notable IFEval score of 95.0. The model features a native 262K-token context window, extensible to 1M tokens via YaRN (RoPE scaling), and is released under the Apache-2.0 license.

Optimized for agentic workflows, Qwen3.5-27B employs an early-fusion architecture that treats visual and textual data as a unified stream for deep cross-modal reasoning. This unified approach allows the model to excel in technical analysis and software engineering, matching GPT-5-mini with a 72.4% score on SWE-bench Verified. While the larger MoE variants in the family lead in raw knowledge benchmarks, the 27B model offers a stable and high-density alternative for structured data extraction and spatial perception, contributing to the Qwen3.5 family’s generational leap in OCR accuracy over the previous Qwen3-VL series.

Frequently Asked Questions

On Roboflow's Vision Evals, GPT-5.6 Terra performed better. It scores higher on 5 of the six vision tasks and averages 72.4% (#12 of 25) against 64.3% (#22 of 25) for Qwen3.5 27B. These scores measure vision capabilities only; either model may perform differently on work outside this benchmark's tasks.

Yes. On the Vision Evals Reasoning benchmark at low effort, GPT-5.6 Terra leads with 59.6% against 31.8%. This is the widest gap between the two models across the benchmark's tasks.

Qwen3.5 27B is cheaper to run. Across Roboflow's Vision Evals task mix it averaged $0.0007 per sample against $0.0044. GPT-5.6 Terra is priced at $1.00 per 1M input tokens and $6.00 per 1M output; Qwen3.5 27B is priced at $0.20 per 1M input tokens and $1.56 per 1M output. Actual costs depend on your image sizes, prompts, and output length.

GPT-5.6 Terra is faster. Across Roboflow's Vision Evals it averaged 7.2s per inference against 7.4s. 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 classification and open prompts in the free Roboflow Playground. You can try it instantly, and a free account unlocks unlimited runs.