Roboflow

Gemini 4 Argon vs Muse Spark 1.1

Gemini 4 Argon and Muse Spark 1.1 score 87.5% vs 74.4% averaged across five Vision Evals tasks (#5 and #24 of 73 models), so Gemini 4 Argon leads overall.

  • Gemini 4 Argon wins all 5 tasks, with its widest lead on object detection (82.5% vs 60.6%).
  • Gemini 4 Argon is cheaper: $0.0058 vs $0.0075 overall average cost per sample.
  • Muse Spark 1.1 is faster: 7.2s vs 13.4s per sample.

Compare Gemini 4 Argon vs Muse Spark 1.1 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
GoogleGemini 4 Argon
Run to compare this model.
MetaMuse Spark 1.1
Run to compare this model.

Gemini 4 Argon vs Muse Spark 1.1 on Vision Evals

Score by taskGemini 4 ArgonMuse Spark 1.1
Object Detection
82.5
60.6
OCR
85.7
68.8
Reasoning
89.4
74.2
Counting
86.0
76.6
Identification
93.8
91.7

Roboflow Vision Evals. Full breakdown below.

Gemini 4 Argon vs Muse Spark 1.1 Comparison Table

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

PropertyGemini 4 ArgonMuse Spark 1.1
OrganizationGoogleMeta
Categoryclosedclosed
Modalitymultimodalmultimodal
Release DateOct 2026Jul 2026
Context Window1.0M1.0M
ParametersUndisclosedUnknown
LicenseProprietaryProprietary
Commercial usecommercial use per vendor termscommercial use per vendor terms
Pricing per 1M tokens
Input $/1M$2.00$1.25
Output $/1M$10.00$4.25
Vision Tasks
Promptable Concept SegmentationDemoNot listed
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
87.5%
74.4%
Avg cost / sample$0.0058$0.0075
Avg speed / sample13.40s7.15s
By task
Object Detection (low)
82.5%
±0.5, Mean of 3 runs, range 81.9 to 83.0
$0.011
60.6%
±1.9, Mean of 3 runs, range 58.4 to 62.1
$0.0097
Object Detection (high)
85.1%
±0.6, Mean of 3 runs, range 84.3 to 85.5
$0.027
60.0%
±0.5, Mean of 3 runs, range 59.6 to 60.7
$0.015
Counting (low)
86.0%
±0.7, Mean of 3 runs, range 85.1 to 86.5
$0.0024
76.6%
±1.3, Mean of 3 runs, range 75.7 to 78.4
$0.0042
Counting (high)
90.5%
±1.4, Mean of 3 runs, range 89.2 to 91.9
$0.010
76.1%
±0.7, Mean of 3 runs, range 75.7 to 77.0
$0.0079
Identification (low)
93.8%
±0.0, Mean of 3 runs, range 93.8 to 93.8
$0.0024
91.7%
±1.6, Mean of 3 runs, range 90.6 to 93.8
$0.0033
Identification (high)
96.9%
±0.0, Mean of 3 runs, range 96.9 to 96.9
$0.0057
89.6%
±1.6, Mean of 3 runs, range 87.5 to 90.6
$0.0065
OCR (low)
85.7%
$0.0051
68.8%
$0.0076
by category
Single value
78.3%
Transcription
95.5%
Structured JSON
92.6%
Text localization
81.5%
Single value
60.4%
Transcription
90.1%
Structured JSON
86.1%
Text localization
33.9%
OCR (high)
88.0%
$0.022
67.5%
$0.015
by category
Single value
80.0%
Transcription
98.8%
Structured JSON
94.8%
Text localization
84.7%
Single value
60.4%
Transcription
81.6%
Structured JSON
85.7%
Text localization
31.4%
Reasoning (low)
89.4%
±1.3, Mean of 3 runs, range 88.1 to 90.7
$0.0024
74.2%
±1.7, Mean of 3 runs, range 72.2 to 75.5
$0.0061
Reasoning (high)
91.4%
±0.7, Mean of 3 runs, range 90.7 to 92.0
$0.013
76.6%
±1.3, Mean of 3 runs, range 75.5 to 78.2
$0.013

Gemini 4 Argon vs Muse Spark 1.1: Overview

Gemini 4 Argon

Gemini 4 Argon is Google DeepMind's frontier Gemini 4 model, built to sustain extended reasoning across long-horizon workflows in software engineering, enterprise knowledge work such as legal and finance, scientific analysis, and defensive cybersecurity. It accepts text, image, video, and document input and returns text, operating over a one million token context window. The model raises the maximum output length to one million tokens, compared with a 64K ceiling on earlier Gemini releases, so a single reasoning trajectory can run for hundreds of thousands of tokens before terminating.

On the visual side, Argon targets multimodal document comprehension, professional chart analysis, and retrieval of specific details from long recordings, reporting 91.7% on LVBench for long-video understanding and 71.6% on Chartography for chart reasoning. Google also reports 77.9% on DeepSWE v1.1, 51.3% on AutomationBench, 68% on CWE-bench v1, and first place on the Vals Index, which aggregates finance, coding, legal, and tax tasks. Following Gemini API conventions, it localizes objects by emitting bounding boxes as [ymin, xmin, ymax, xmax] integers normalized to a 0 to 1000 range, which supports prompt driven detection and grounding alongside captioning, document parsing, and visual question answering. Google describes adversarial training and automated red teaming for indirect prompt injection resistance, plus chain-of-thought and action monitors that halt execution when behavior drifts outside user intent.

Muse Spark 1.1

Muse Spark 1.1 is a natively multimodal reasoning model from Meta Superintelligence Labs, released on July 9, 2026, as a significant upgrade to the original Muse Spark. The model accepts text, image, video, PDF, and audio as input and produces text output. It operates with a 1-million-token context window (1,048,576 tokens per the Meta Model API documentation) and is designed specifically for agentic tasks that require planning, tool use, computer use, and multi-agent orchestration. The model runs in a "Thinking" mode, where adjustable reasoning effort is applied before generating a response. It can function both as a main agent gathering context, forming plans, and delegating to parallel subagents and as a subagent that adheres to assigned tasks and escalates when needed. It is trained to decide autonomously when to write automation scripts versus interact directly with a user interface.

Muse Spark 1.1 supports a range of multimodal capabilities including visual perception, image and video captioning, visual-to-code generation, and document analysis. The model was evaluated under Meta's Advanced AI Scaling Framework across frontier risk categories including chemical and biological threats, cybersecurity, and loss-of-control scenarios. Parameter count, architecture details, and training data composition are not publicly disclosed. The model is proprietary and closed-weight, accessible to consumers through the Meta AI app and to developers via the Meta Model API, which launched in public preview alongside this release.