Roboflow

GPT-5.6 Sol vs Muse Spark 1.2

Muse Spark 1.2 and GPT-5.6 Sol score 73.8% vs 72.4% averaged across five Vision Evals tasks (#26 and #27 of 73 models), so Muse Spark 1.2 leads overall.

  • Muse Spark 1.2 wins 3 of the 5 tasks, but the widest gap is object detection, where GPT-5.6 Sol leads 68.4% vs 59.0%.
  • Muse Spark 1.2 is about half the cost: $0.0083 vs $0.019 overall average cost per sample.
  • Muse Spark 1.2 is faster: 11.5s vs 12.2s per sample.

Compare GPT-5.6 Sol vs Muse Spark 1.2 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 Sol
Run to compare this model.
MetaMuse Spark 1.2
Run to compare this model.

GPT-5.6 Sol vs Muse Spark 1.2 on Vision Evals

Score by taskGPT-5.6 SolMuse Spark 1.2
Object Detection
68.4
59.0
Reasoning
66.0
75.1
OCR
63.6
68.7
Counting
74.3
76.6
Identification
89.6
89.6

Roboflow Vision Evals. Full breakdown below.

GPT-5.6 Sol vs Muse Spark 1.2 Comparison Table

Evals updated November 19, 2026Pricing updated November 22, 2026

PropertyGPT-5.6 SolMuse Spark 1.2
OrganizationOpenAIMeta
Categoryclosedclosed
Modalitymultimodalmultimodal
Release DateJul 2026Aug 2026
Context Window1.5M1.0M
ParametersUnknownUnknown
LicenseProprietaryProprietary
Commercial usecommercial use per vendor termscommercial use per vendor terms
Pricing per 1M tokens
Input $/1M$4.00$1.25
Output $/1M$20.00$4.25
Vision Tasks
Both support 10 more tasks
Model Features
Both support 3 more features
Vision Evalsground-truth scores across 5 vision tasks, pooled at low effort
Overall
72.4%
73.8%
Avg cost / sample$0.019$0.0083
Avg speed / sample12.16s11.51s
By task
Object Detection (low)
68.4%
±0.7, Mean of 3 runs, range 67.9 to 69.3
$0.031
59.0%
±1.0, Mean of 3 runs, range 58.1 to 60.2
$0.0096
Object Detection (high)
68.4%
±0.8, Mean of 3 runs, range 67.7 to 69.3
$0.071
60.5%
±0.3, Mean of 3 runs, range 60.2 to 60.7
$0.014
Counting (low)
74.3%
±1.4, Mean of 3 runs, range 73.0 to 75.7
$0.0098
76.6%
±2.7, Mean of 3 runs, range 74.3 to 79.7
$0.0050
Counting (high)
76.1%
±2.0, Mean of 3 runs, range 74.3 to 78.4
$0.016
75.2%
±2.0, Mean of 3 runs, range 73.0 to 77.0
$0.0082
Identification (low)
89.6%
±4.7, Mean of 3 runs, range 84.4 to 93.8
$0.0056
89.6%
±1.6, Mean of 3 runs, range 87.5 to 90.6
$0.0038
Identification (high)
92.7%
±1.6, Mean of 3 runs, range 90.6 to 93.8
$0.0060
87.5%
±0.0, Mean of 3 runs, range 87.5 to 87.5
$0.0062
OCR (low)
63.6%
$0.019
68.7%
$0.0086
by category
Single value
43.5%
Transcription
91.7%
Structured JSON
82.0%
Text localization
52.2%
Single value
60.4%
Transcription
89.3%
Structured JSON
87.0%
Text localization
31.5%
OCR (high)
65.3%
$0.048
68.6%
$0.014
by category
Single value
46.1%
Transcription
92.3%
Structured JSON
83.9%
Text localization
51.6%
Single value
61.7%
Transcription
85.1%
Structured JSON
86.7%
Text localization
30.7%
Reasoning (low)
66.0%
±2.6, Mean of 3 runs, range 63.6 to 68.9
$0.0086
75.1%
±0.3, Mean of 3 runs, range 74.8 to 75.5
$0.0073
Reasoning (high)
71.7%
±1.3, Mean of 3 runs, range 70.2 to 72.8
$0.012
75.7%
±0.3, Mean of 3 runs, range 75.5 to 76.2
$0.012

GPT-5.6 Sol vs Muse Spark 1.2: Overview

GPT-5.6 Sol

GPT-5.6 Sol is the flagship model in OpenAI's GPT-5.6 family, which also includes Terra (a balanced everyday-work tier) and Luna (a fast, cost-efficient tier). Sol is designed for demanding reasoning, long-horizon agentic workflows, software engineering, computer use, scientific research, and cybersecurity tasks. It introduces two new capability modes: a "max" reasoning effort setting that allocates additional compute time for difficult problems, and an "ultra" mode that coordinates multiple subagents in parallel to accelerate complex, multi-step work. The model supports native multimodal input, allowing it to process screenshots, diagrams, charts, documents, and photographs alongside text. A reported context window of approximately 1.5 million tokens enables processing of large codebases, lengthy research documents, and extended agentic sessions.

GPT-5.6 Sol was announced on June 26, 2026, initially in a limited preview for trusted partners, and reached general availability on July 9, 2026. On the Agents' Last Exam benchmark, which evaluates long-running professional workflows across 55 fields, Sol scores 53.6. The model also demonstrates gains in life sciences evaluations, including long-horizon genomics and quantitative biology analyses. OpenAI paired the release with its most extensive safety evaluation to date, combining human red teaming with large-scale automated testing, and classified Sol as High capability in both cybersecurity and biological risk under its Preparedness Framework, though it does not cross the Critical threshold in either category.

Muse Spark 1.2

Muse Spark 1.2 is a proprietary multimodal reasoning model from Meta Superintelligence Labs, released as a coding-focused update to Muse Spark 1.1. It accepts text, images, video, audio, and PDF documents and returns text, with a context window of roughly one million tokens that allows whole repositories, long documents, and extended agent trajectories to be held in a single request. The model thinks before answering, and the amount of reasoning effort it spends is configurable per request. Alongside its visual and document understanding, it supports structured output and parallel function calling, and it is designed to operate either as a planning agent that delegates work or as a subagent executing tasks in parallel.

Training for version 1.2 scaled up compute on coding tasks and widened the diversity of training environments, concentrating on long-horizon work such as whole-repository generation, large end-to-end projects, and automated research. Part of the training data was self-generated, with Muse Spark 1.1 producing coding environments and instruction-following templates and grading candidate solutions against them. The model was co-trained with the Muse Code terminal agent, incorporating rejection-sampled harness trajectories and that toolset. Multimodal use cases documented for the family include visual-to-code generation and detailed image and video captioning.