ZMIME
Comparison · 3 models · Updated Nov 22, 2026

Muse Spark 1.3 vs Gemini 3.8 Flash vs GLM-5.3

Too close to call on our weighted score (Gemini 3.8 Flash 74, Muse Spark 1.3 73, GLM-5.3 64). The right pick depends on what you value most.Too close to call on our weighted score (Gemini 3.8 Flash 86, Muse Spark 1.3 86, GLM-5.3 74). The right pick depends on what you value most.

  1. Meta

    Muse Spark 1.3

    Released Sep 2, 2026Latest

    73/100
    • ECI158.2
    • Price$1.25 / $4.25
    • Context1.05M
  2. Google

    Gemini 3.8 Flash

    Released Sep 2, 2026Latest

    74/100
    • ECI158.8
    • Price$0.75 / $3.75
    • Context1.05M
  3. Z.ai (Zhipu)

    GLM-5.3

    Released Aug 14, 2026Latest

    64/100
    • ECI157.1
    • Price$1.40 / $4.40
    • Context1M
  4. Meta

    Muse Spark 1.3

    Released Sep 2, 2026Latest

    86/100
    • ECI158.2
    • Price$1.25 / $4.25
    • Context1.05M
  5. Google

    Gemini 3.8 Flash

    Released Sep 2, 2026Latest

    86/100
    • ECI158.8
    • Price$0.75 / $3.75
    • Context1.05M
  6. Z.ai (Zhipu)

    GLM-5.3

    Released Aug 14, 2026Latest

    74/100
    • ECI157.1
    • Price$1.40 / $4.40
    • Context1M
01 — Verdict

Too close to call

It is close. Our weighted score puts them within 2 points (Gemini 3.8 Flash 74/100, Muse Spark 1.3 73/100, GLM-5.3 64/100), so choose by what matters most for your work: Gemini 3.8 Flash for raw capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.It is close. Our weighted score puts them within a point (Gemini 3.8 Flash 86/100, Muse Spark 1.3 86/100, GLM-5.3 74/100), so choose by what matters most for your work: Gemini 3.8 Flash for raw capability. The score weighs capability 67%, inputs & features 20%, context window 13%.

  • CapabilityGemini 3.8 FlashCapabilities Index (ECI): Gemini 3.8 Flash 158.8 · Muse Spark 1.3 158.2 · GLM-5.3 157.1
  • Lowest priceGemini 3.8 FlashGemini 3.8 Flash $1.50 · Muse Spark 1.3 $2.00 · GLM-5.3 $2.15 per 1M tokens (3:1 blend)
  • Longest contextMuse Spark 1.3 and Gemini 3.8 FlashMuse Spark 1.3 1,048,576 · Gemini 3.8 Flash 1,048,576 · GLM-5.3 1,000,000 tokens
  • Widest inputsMuse Spark 1.3 and Gemini 3.8 FlashMuse Spark 1.3: Text, Images, PDFs, Audio, Video · Gemini 3.8 Flash: Text, Images, PDFs, Audio, Video · GLM-5.3: Text
  • Self-hostingGLM-5.3Publishes downloadable weights
How the score is built
MeasureWeightMuse Spark 1.3Gemini 3.8 FlashGLM-5.3
CapabilityCapabilities Index (ECI)50%868685
Price25%344033
Inputs & features15%10010045
Context window10%656564
Overall100%73/10074/10064/100
How the score is built
MeasureWeightMuse Spark 1.3Gemini 3.8 FlashGLM-5.3
CapabilityCapabilities Index (ECI)67%868685
Inputs & features20%10010045
Context window13%656564
Overall100%86/10086/10074/100
02 — Side by side

Every spec in one table

Highlighted cells lead their row. Unavailable specifications are marked “Not published”. Publisher results include their evaluation settings and source.

Muse Spark 1.3 vs Gemini 3.8 Flash vs GLM-5.3 specifications side by side
SpecificationMuse Spark 1.3MetaGemini 3.8 FlashGoogleGLM-5.3Z.ai (Zhipu)
Capability
Capabilities Index (ECI)158.2158.8 (best)157.1
ECI rank#28 of 159#25 of 159 (best)#34 of 159
ECI confidence interval156.2–160.5156.8–161.0155.0–159.4
ECI evaluation dateSep 2, 2026Sep 2, 2026Aug 14, 2026
Documented capabilitiesMuse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflowsFlagship GLM model for long-horizon coding, agents, and complex project delivery
Model parametersNot publishedNot publishedNot published
Output modalitiesTextTextText
GPQA DiamondGraduate-level science questionsNot published95.4%high (best)90.9%max
FrontierMath Tiers 1–3Research-level mathematics74.4%xhigh (best)68.4%high68.8%max
OTIS Mock AIME 2024–2025Competition mathematics99.2%xhigh (best)98.9%high91.1%max
SimpleQA VerifiedShort factual questionsNot published69.7%high (best)41.0%max
Publisher-reported benchmarks
GDPval-AA 2max effort1754EloReported resultSep 2, 2026Not published1769EloReported result
JobBenchmax effort64.9mean rubric scoreReported resultSep 2, 2026Not publishedNot published
OSWorld 2.0 · 08.08max effort66.9mean partial scoreReported resultSep 2, 2026Not publishedNot published
OSWorld 2.0 · 08.08max effort32binary completion rateReported resultSep 2, 2026Not publishedNot published
DeepSearchQAmax effort90.3F1Reported resultSep 2, 2026Not publishedNot published
Agentic IF Index · Meta internalmax effort57.8scoreReported resultSep 2, 2026Not publishedNot published
AutomationBench · public v3 task setmax effort49.6pass@1Reported resultSep 2, 2026Not publishedNot published
MRCR 2 · 8-needle, 256K-512Kmax effort98.5mean sequence-match ratioReported resultSep 2, 2026Not publishedNot published
MRCR 2 · 8-needle, 512K-1Mmax effort98.1mean sequence-match ratioReported resultSep 2, 2026Not publishedNot published
DeepSWE 1.1max effort75.4pass@1Reported resultSep 2, 2026Not publishedNot published
SWE-Atlas Codebase QnAmax effort59.4pass@1Reported resultSep 2, 2026Not publishedNot published
Terminal-Bench 2.1max effort88.8pass@1Reported resultSep 2, 2026Not published88.2pass@1Reported result
DeepSWE 1.1high thinkingNot published73.7pass@1Reported resultNot published
GDPval-AA 2Publisher-reported resultNot published1545EloReported resultNot published
Finance Agent 2Publisher-reported resultNot published61.4pass@1Reported resultNot published
Harvey Legal Agent BenchmarkPublisher-reported resultNot published10all-pass rateReported resultNot published
Terminal-Bench 2.1Publisher-reported resultNot published89.4pass@1Reported resultNot published
Terminal-Bench 4.0Publisher-reported resultNot published19.1pass@1Reported resultNot published
GDP.PDFPublisher-reported resultNot published35all-pass rateReported resultNot published
CharXiv Reasoningwithout toolsNot published86.2pass@1Reported resultNot published
LVBenchagenticNot published87.8pass@1Reported resultNot published
LVBenchstatic; 1024 frames; without toolsNot published87.1pass@1Reported resultNot published
HLE-Verified · 1811 verified and revised itemsPublisher-reported resultNot published54.9accuracyReported resultNot published
OSWorld 2.0 · before 2026-08-08 patchbatched tool calls; best of 3 runs; 500 stepsNot published59partial scoreReported resultNot published
BioMysteryBench · human-solvablewith toolsNot published88.8pass@1Reported resultNot published
BioMysteryBench · human-difficultwith toolsNot published56.5pass@1Reported resultNot published
LAB-Bench 2 · 11 subtaskswith toolsNot published86.2macro-averageReported resultNot published
Terminal-Bench 3.0max effortNot publishedNot published28.3avg@3Reported result
DeepSWE 1.1max effort; 6h timeout; 400K contextNot publishedNot published66.9resolvedReported result
NL2Repomax effortNot publishedNot published58scoreReported result
ProgramBenchmax effortNot publishedNot published19almost solvedReported result
FrontierSWEmax effortNot publishedNot published78.1dominance scoreReported resultAug 14, 2026
SWE-Marathon 1.1max effort; modified anti-cheat checksNot publishedNot published42.5scoreReported result
PostTrainBenchmax effort; 3 runs; modified external-API checksNot publishedNot published39.8weighted averageReported result
CyberGymmax effortNot publishedNot published84.5pass@1Reported result
ExploitGym · 869 tasksmax effort; 2h rescaled timeoutNot publishedNot published105tasks solvedReported result
ExploitGym · 869 tasksmax effort; 6h rescaled timeoutNot publishedNot published130tasks solvedReported result
ExploitBenchmax effort; union over 3 revisionsNot publishedNot published54.4average coverageReported result
Toolathlon-Verifiedmax effortNot publishedNot published73pass@1Reported result
AutomationBench 1.0.6max effortNot publishedNot published48.2pass@1Reported result
Agents' Last Exammax effort; CLINot publishedNot published28.5scoreReported result
Humanity's Last Exammax effort; with toolsNot publishedNot published62.5accuracyReported result
Price per million tokens
Input$1.25$0.75 (best)$1.40
Output$4.25$3.75 (best)$4.40
Cached input$0.15$0.075 (best)$0.26
Blended (3:1)$2.00$1.50 (best)$2.15
Long-context rateSame rateSame rateSame rate
Price sourceOfficial Meta APIOfficial Google APIOfficial Z.AI API
Limits
Context window1,048,576 tokens (best)1,048,576 tokens (best)1,000,000 tokens
Max inputNot publishedNot publishedNot published
Max output131,072 tokens (best)65,536 tokens131,072 tokens (best)
Inputs and features
TextYesYesYes
ImagesYesYesNo
PDFsYesYesNo
AudioYesYesNo
VideoYesYesNo
ReasoningYesminimal · low · medium · high · xhigh · maxYeslow · medium · highYeslow · high · max
Tool callingYesYesYes
Structured outputYesYesYes
File attachmentsYesYesNo
Temperature controlYesYesYes
Reasoning controleffortefforteffort
Availability
WeightsProprietaryProprietaryOpen
API model IDmuse-spark-1.3gemini-3.8-flashglm-5.3
API providers1842123 (best)
ReleasedSep 2, 2026Sep 2, 2026Aug 14, 2026
Knowledge cutoffNot publishedNot publishedNot published
Publisher documentationPublisher documentationPublisher documentationPublisher documentation · Model weights
Source metadata updatedSep 2, 2026Sep 2, 2026Aug 14, 2026
Data sourcesmodels.dev · Epoch AImodels.dev · Epoch AImodels.dev · Epoch AI
Catalog checkedNov 22, 2026Nov 22, 2026Nov 22, 2026
03 — Cost

What would a month cost?

Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.

  • Muse Spark 1.3$21.00
  • Gemini 3.8 Flash$15.00
  • GLM-5.3$22.80
04 — Questions

Which should you choose?

Which is better: Muse Spark 1.3, Gemini 3.8 Flash or GLM-5.3?

It is close. Our weighted score puts them within 2 points (Gemini 3.8 Flash 74/100, Muse Spark 1.3 73/100, GLM-5.3 64/100), so choose by what matters most for your work: Gemini 3.8 Flash for raw capability. The score weighs capability 50%, price 25%, inputs & features 15%, context window 10%.It is close. Our weighted score puts them within a point (Gemini 3.8 Flash 86/100, Muse Spark 1.3 86/100, GLM-5.3 74/100), so choose by what matters most for your work: Gemini 3.8 Flash for raw capability. The score weighs capability 67%, inputs & features 20%, context window 13%.

Which is cheaper, Muse Spark 1.3, Gemini 3.8 Flash or GLM-5.3?

Gemini 3.8 Flash is cheaper at $0.75 input / $3.75 output per million tokens (official Google API price). Muse Spark 1.3 costs $1.25 input / $4.25 output per million tokens (official Meta API price); GLM-5.3 costs $1.40 input / $4.40 output per million tokens (official Z.AI API price). At a typical mix of three input tokens to one output token, that is $1.50 per million tokens for Gemini 3.8 Flash versus $2.00 for Muse Spark 1.3 (1.3× as much) and $2.15 for GLM-5.3 (1.4× as much).

Which scores higher on benchmarks?

Gemini 3.8 Flash scores higher on the Capabilities Index (ECI): Gemini 3.8 Flash 158.8 (#25 of 159), Muse Spark 1.3 158.2 (#28 of 159) and GLM-5.3 157.1 (#34 of 159). The confidence ranges of the top two overlap (156.8–161.0 vs 156.2–160.5), so treat the gap as small. On individual benchmarks: FrontierMath Tiers 1–3 — Muse Spark 1.3 74.4%, GLM-5.3 68.8%, Gemini 3.8 Flash 68.4%; OTIS Mock AIME 2024–2025 — Muse Spark 1.3 99.2%, Gemini 3.8 Flash 98.9%, GLM-5.3 91.1%.

Which is better for coding?

There are no published SWE-bench Verified results for Muse Spark 1.3, Gemini 3.8 Flash and GLM-5.3 yet, so there is no like-for-like coding score. On overall capability, Gemini 3.8 Flash leads, which tends to carry over to coding, but test on your own codebase. All three support tool calling for agent workflows.

Which has the bigger context window?

Muse Spark 1.3 and Gemini 3.8 Flash have the largest context windows (1,048,576 and 1,048,576 tokens), against 1,000,000 for GLM-5.3. Maximum output per response: Muse Spark 1.3 up to 131,072, Gemini 3.8 Flash up to 65,536, GLM-5.3 up to 131,072 tokens.

Which can read images, PDFs, audio or video?

Muse Spark 1.3 accepts text, images, PDFs, audio and video; Gemini 3.8 Flash accepts text, images, PDFs, audio and video; GLM-5.3 accepts text. Muse Spark 1.3 handles the widest range of inputs.

Are any of these open source?

GLM-5.3 publishes its weights and can be self-hosted; Muse Spark 1.3 and Gemini 3.8 Flash is proprietary.

Which is newer?

Muse Spark 1.3 is the newest, released Sep 2, 2026. Gemini 3.8 Flash came out Sep 2, 2026; GLM-5.3 came out Aug 14, 2026.

How do you decide the winner?

This rating uses capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 22, 2026.This rating uses capability (67%, independent benchmark results); inputs & features (20%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (13%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 22, 2026.