ModelBench
Meta logo
Models/Meta

Meta: Muse Spark 1.3

Muse 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.

Source-linked muse Closed weights Released 2026-09-02
API record Report
InputT
OutputT
Input price$1.25/M
Output price$4.25/M
Context1.04858M
Max output131.072K
Providers9
Inference availability

Providers

Provider-specific identifiers, limits, and listed prices per million tokens. Every row links back to the provider's own documentation.

Report a price
Providers offering Muse Spark 1.3
ProviderProvider model IDContextMax outputInputOutputCache readCapabilitiesDocs
Vercel AI Gateway meta/muse-spark-1.3 1.04858M 1.04858M $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
Kilo Gateway meta/muse-spark-1.3 1.04858M 943.718K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
OpenRouter meta/muse-spark-1.3 1.04858M 943.718K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
NanoGPT meta/muse-spark-1.3 1.04858M 943.718K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
LLM Gateway meta/muse-spark-1.3 1.04858M 1.04858M $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
DevPass (LLM Gateway) muse-spark-1.3 1.04858M 131.072K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
Meta muse-spark-1.3 1.04858M 131.072K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗
EmpirioLabs AI muse-spark-1-3 1.04858M 131.072K $1.25 $4.25 $1 ReasoningToolsJSON Docs ↗
OpenCode Zen muse-spark-1.3 1.04858M 131.072K $1.25 $4.25 $0.15 ReasoningToolsJSON Docs ↗

Capability badges appear only where the provider catalog explicitly lists support. A blank cell means the source is silent, not that the feature is absent.

Listed rates

Price across providers

Input price per million tokens as published by each provider. Bars are drawn from listed rates only — no traffic weighting, since the catalog observes no requests.

Lowest input $1.25/M

Across 9 priced providers

Median input $1.25/M

Midpoint of listed rates

Highest input $1.25/M

Same as the lowest listed rate

Output range $4.25 – $4.25

Per million output tokens

Vercel AI Gateway $1.25/M
Kilo Gateway $1.25/M
OpenRouter $1.25/M
NanoGPT $1.25/M
LLM Gateway $1.25/M
DevPass (LLM Gateway) $1.25/M
Meta $1.25/M
EmpirioLabs AI $1.25/M
OpenCode Zen $1.25/M
Cost calculator

Estimate a workload

$0.00
Excludes taxes, non-token charges, and tiered discounts.

9 providers list the identical $1.25 input rate, so price alone will not separate them — compare context limits, max output, and capabilities above.

Specification

Capabilities

Recorded from the source catalog and provider listings.

✓ Reasoning Yes
✓ Tool calling Yes
✓ Structured output Yes
✓ Attachments Yes
✓ Vision input Yes
× Open weights No
Creator
Meta
Model family
muse
Knowledge cutoff
Not documented
License
Not documented
Release date
2026-09-02
Model ID
meta/muse-spark-1.3
Published evaluations

Benchmarks

Every result stays attached to its source, version, metric and harness. Scores from different versions are never merged, and the profile below plots each benchmark against its own population rather than on a shared scale.

Benchmark registry
Benchmark profile
Design Arena: w…Design Arena: website Design Arena: c…Design Arena: codecategories Design Arena: g…Design Arena: gamedev Design Arena: d…Design Arena: dataviz Design Arena: u…Design Arena: uicomponent Design Arena: 3dDesign Arena: 3d Design Arena: f…Design Arena: fullstack Design Arena: w…Design Arena: webapps This model — Design Arena: website: 1359 elo (99.7th percentile) This model — Design Arena: codecategories: 1366 elo (98.5th percentile) This model — Design Arena: gamedev: 1365 elo (94.9th percentile) This model — Design Arena: dataviz: 1327 elo (93.6th percentile) This model — Design Arena: uicomponent: 1394 elo (99.7th percentile) This model — Design Arena: 3d: 1419 elo (97.1th percentile) This model — Design Arena: fullstack: 1314 elo (85.8th percentile) This model — Design Arena: webapps: 1311 elo (94.8th percentile)

Each axis is this model's percentile among the 8 benchmarks it has published results for, measured against every other model with a score on that same benchmark. Percentiles are used because benchmarks do not share a scale — a 60 on one is not a 60 on another. Hover any point for the raw score.

Design Arena: website elo Source ↗
Design Arena: website #1 of 175
Design Arena: codecategories elo Source ↗
Design Arena: codecategories #3 of 167
Design Arena: gamedev elo Source ↗
Design Arena: gamedev #9 of 166
Design Arena: dataviz elo Source ↗
Design Arena: dataviz #11 of 165
Design Arena: uicomponent elo Source ↗
Design Arena: uicomponent #1 of 161
Design Arena: 3d elo Source ↗
Design Arena: 3d #5 of 157
Design Arena: fullstack elo Source ↗
Design Arena: fullstack #10 of 67
Design Arena: webapps elo Source ↗
Design Arena: webapps #4 of 67
Design Arena: python-pptxslides elo Source ↗
Design Arena: python-pptxslides #19 of 47

Each strip shows every published score for that benchmark, with this model marked. The lighter shape behind the ticks is the density of results, and the dashed line is the median.

Cost against capability

Price and performance

Listed input price plotted against Design Arena: website, the benchmark with the widest published coverage that this model appears in.

0 720 1441 $0.1 $1 $10 Anthropic: Claude Fable 5.1 (batch) — $5/M, score 1322.0 Claude Opus 5 — $5/M, score 1320.0 Mistral: Ministral 3 8B 2512 (batch) — $0.075/M, score 1077.0 DeepSeek Chat — $0.147/M, score 1132.0 OpenAI: GPT-4o (batch) — $1.25/M, score 843.0 Anthropic: Claude Opus 4.7 — $5/M, score 1307.0 Inception: Mercury 2 — $0.25/M, score 1035.0 Gemini 3.1 Flash Lite Preview — $0.25/M, score 1095.0 Z.ai: GLM 5 — $0.6/M, score 1260.0 Mistral: Mistral Large 3 2512 (batch) — $0.25/M, score 1175.0 Anthropic: Claude Sonnet 4.5 — $3/M, score 1202.0 Muse Spark 1.3 — $1.25/M, score 1359.0 DeepSeek: DeepSeek V3.2 Exp — $0.27/M, score 1191.0 Anthropic: Claude Opus 4 — $15/M, score 1177.0 Qwen: Qwen3 235B A22B — $0.455/M, score 1043.0 OpenAI: GPT-4.1 (batch) — $1/M, score 1051.0 Gemini 3.5 Flash — $1.5/M, score 1275.0 Z.ai: GLM 5.3 Flash (batch) — $0.075/M, score 1291.0 DeepSeek V4 Flash — $0.15/M, score 1220.0 DeepSeek V4 Pro — $0.435/M, score 1247.0 GPT-4.1 mini — $0.4/M, score 1010.0 GPT OSS 120B — $0.03/M, score 980.0 GPT-5.2 — $1.75/M, score 1207.0 GPT-5 — $1.25/M, score 1197.0 Claude Sonnet 5 — $2/M, score 1289.0 Mistral: Ministral 3 8B 2512 — $0.15/M, score 1077.0 Z.ai: GLM 5.3 (batch) — $0.7/M, score 1319.0 Z.ai: GLM 5.1 — $0.966/M, score 1290.0 Gemini 3.7 Flash — $0.75/M, score 1315.0 Z.ai: GLM 5.3 — $1.4/M, score 1319.0 Google: Gemini 3.1 Pro Preview (batch) — $1/M, score 1265.0 Mistral: Codestral 2508 (batch) — $0.15/M, score 1025.0 Google: Gemini 2.5 Pro (batch) — $0.625/M, score 1179.0 OpenAI: GPT-5.5 (batch) — $2.5/M, score 1269.0 GPT-5.3 Codex — $1.75/M, score 1175.0 Gemini 3.6 Flash — $0.75/M, score 1312.0 Z.ai: GLM 4.7 Flash — $0.061/M, score 1206.0 Claude Fable 5 — $10/M, score 1310.0 Z.ai: GLM 5.3 Flash — $0.15/M, score 1291.0 Mistral: Codestral 2508 — $0.3/M, score 1025.0 MiniMax: MiniMax M3 — $0.3/M, score 1271.0 Google: Gemini 2.5 Flash (batch) — $0.15/M, score 1127.0 Anthropic: Claude Fable 5.1 — $10/M, score 1322.0 Z.ai: GLM 5V Turbo — $1.2/M, score 1243.0 MoonshotAI: Kimi K2.7 Code (batch) — $0.95/M, score 1290.0 OpenAI: o3 (batch) — $1/M, score 1048.0 Inkling — $1.87/M, score 1226.0 OpenAI: gpt-oss-20b (batch) — $0.05/M, score 865.0 Muse Spark 1.1 — $1.25/M, score 1282.0 MiMo-V2.5 — $0.14/M, score 1279.0 MoonshotAI: Kimi K2 0711 — $0.57/M, score 1063.0 Kimi K2 Thinking — $0.4/M, score 1125.0 Qwen: Qwen3.8 Max (0803) — $2/M, score 1295.0 Mistral: Mistral Medium 3.1 (batch) — $0.2/M, score 1145.0 MiMo-V2.5-Pro — $0.435/M, score 1285.0 GPT-5.1 — $1.25/M, score 1199.0 Hy3 — $0.066/M, score 1194.0 Google: Gemini 3.8 Flash (batch) — $0.375/M, score 1315.0 OpenAI: GPT-5 Mini (batch) — $0.125/M, score 1137.0 Gemini 2.5 Pro — $1.25/M, score 1179.0 GPT-5.5 — $5/M, score 1269.0 MoonshotAI: Kimi K3 (batch) — $3/M, score 1354.0 Anthropic: Claude Sonnet 5 (batch) — $1/M, score 1289.0 DeepSeek V3.2 — $0.18/M, score 1187.0 Anthropic: Claude Fable 5 (batch) — $5/M, score 1310.0 Anthropic: Claude Opus 4.8 — $5/M, score 1267.0 Thinking Machines: Inkling (batch) — $1/M, score 1226.0 SpaceXAI: Grok 4.6 — $2/M, score 1306.0 DeepSeek: DeepSeek V3.1 Terminus — $0.27/M, score 1199.0 GPT-5 Mini — $0.25/M, score 1137.0 Claude Opus 5 (batch) — $2.5/M, score 1320.0 Anthropic: Claude Opus 4.8 (batch) — $2.5/M, score 1267.0 SpaceXAI: Grok 4.3 — $1.25/M, score 1206.0 Google: Gemini 3.5 Flash (batch) — $0.75/M, score 1275.0 o3 — $2/M, score 1048.0 OpenAI: GPT-5.4 (batch) — $1.25/M, score 1232.0 OpenAI: GPT-5.2 (batch) — $0.875/M, score 1207.0 Mistral: Mistral Large 3 2512 — $0.5/M, score 1175.0 GPT-5.1 Codex mini — $0.22/M, score 1123.0 MiniMax: MiniMax M2 — $0.255/M, score 1155.0 Qwen: Qwen3 Max — $0.78/M, score 1131.0 OpenAI: GPT-5 Nano (batch) — $0.025/M, score 1114.0 Anthropic: Claude Opus 4.1 (batch) — $7.5/M, score 1189.0 Mistral: Mistral Medium 3 — $0.4/M, score 1091.0 o4-mini — $1.1/M, score 998.0 GPT-4.1 — $2/M, score 1051.0 Z.ai: GLM 5.2 (batch) — $0.7/M, score 1307.0 Google: Gemini 3.6 Flash (batch) — $0.375/M, score 1312.0 Gemini 3 Flash Preview — $0.5/M, score 1208.0 MiniMax: MiniMax M3 (batch) — $0.3/M, score 1271.0 Gemini 3.8 Flash — $0.75/M, score 1315.0 NVIDIA: Nemotron 3 Ultra (batch) — $0.6/M, score 1144.0 OpenAI: o4 Mini (batch) — $0.55/M, score 998.0 MiniMax: MiniMax M2.7 — $0.3/M, score 1258.0 SpaceXAI: Grok 4.5 — $2/M, score 1296.0 Mistral: Mistral Medium 3.1 — $0.4/M, score 1145.0 Gemini 2.5 Flash — $0.3/M, score 1127.0 OpenAI: gpt-oss-120b (batch) — $0.15/M, score 980.0 OpenAI: GPT-4.1 Mini (batch) — $0.2/M, score 1010.0 Qwen: Qwen3 Coder 480B A35B — $0.3/M, score 1171.0 Solar Pro 4 — $0.3/M, score 1188.0 Gemini 3.1 Pro Preview — $2/M, score 1265.0 Mistral: Mistral Small 3.2 24B — $0.075/M, score 908.0 Anthropic: Claude Opus 4.7 (batch) — $2.5/M, score 1307.0 OpenAI: GPT-4.1 Nano (batch) — $0.05/M, score 985.0 OpenAI: GPT-5.1 (batch) — $0.625/M, score 1199.0 Anthropic: Claude Opus 4.1 — $15/M, score 1189.0 Kimi K2.7 Code — $0.95/M, score 1279.0 Kimi K3 — $3/M, score 1354.0 GPT-5.4 — $2.5/M, score 1232.0 Anthropic: Claude Sonnet 4.6 — $3/M, score 1297.0 Anthropic: Claude Sonnet 4.6 (batch) — $1.5/M, score 1297.0 Google: Gemini 3.7 Flash (batch) — $0.375/M, score 1315.0 Muse Spark 1.2 — $1.25/M, score 1322.0 Anthropic: Claude Opus 4.5 (batch) — $2.5/M, score 1259.0 DeepSeek: DeepSeek V4 Flash 0731 (batch) — $0.11/M, score 1251.0 Qwen: Qwen3.7 Plus — $0.32/M, score 1282.0 SpaceXAI: Grok 4.3 (batch) — $1/M, score 1206.0 Kimi K2.6 — $0.95/M, score 1282.0 Nemotron 3 Ultra 550B A55B — $0.5/M, score 1144.0 Anthropic: Claude Sonnet 4 — $3/M, score 1158.0 SpaceXAI: Grok 4.20 — $1.25/M, score 1242.0 Anthropic: Claude Opus 4.6 (batch) — $2.5/M, score 1304.0 Qwen: Qwen3.6 Plus — $0.325/M, score 1253.0 GPT-5.1 Codex — $1.07/M, score 1174.0 Anthropic: Claude Haiku 4.5 (batch) — $0.5/M, score 1135.0 Z.ai: GLM 4.6 — $0.43/M, score 1187.0 Anthropic: Claude Sonnet 4.5 (batch) — $1.5/M, score 1202.0 GPT-4o — $2.5/M, score 843.0 Google: Gemini 3 Flash Preview (batch) — $0.25/M, score 1208.0 OpenAI: GPT-5 (batch) — $0.625/M, score 1197.0 DeepSeek V4 Flash 0731 — $0.05/M, score 1251.0 GPT-4.1 nano — $0.1/M, score 985.0 GPT OSS 20B — $0.02/M, score 865.0 Kimi K2.5 — $0.3/M, score 1261.0 Trinity Large Thinking — $0.25/M, score 1149.0 MoonshotAI: Kimi K2 0905 — $0.6/M, score 1120.0 Qwen: Qwen3 30B A3B Thinking 2507 — $0.2/M, score 943.0 Z.ai: GLM 5.2 — $0.966/M, score 1307.0 Qwen: Qwen3.7 Max — $1.475/M, score 1285.0 Qwen: Qwen3.5 Plus 2026-02-15 — $0.26/M, score 1200.0 MiniMax: MiniMax M2.1 — $0.3/M, score 1214.0 Z.ai: GLM 4.7 — $0.4/M, score 1238.0 Mistral: Ministral 3 3B 2512 — $0.1/M, score 1040.0 Amazon: Nova Premier 1.0 — $2.5/M, score 847.0 DeepSeek: DeepSeek V3.1 — $0.25/M, score 1135.0 Z.ai: GLM 4.5 — $0.6/M, score 1182.0 Meta: Llama 4 Maverick — $0.2/M, score 883.0 Nex AGI: Nex-N2-Pro — $0.25/M, score 1236.0 Z.ai: GLM 5 Turbo — $1.2/M, score 1278.0 Qwen: Qwen3.5 397B A17B — $0.55/M, score 1203.0 MiniMax: MiniMax M2.5 — $0.3/M, score 1235.0 Anthropic: Claude Opus 4.6 — $5/M, score 1304.0 Mistral: Ministral 3 14B 2512 — $0.2/M, score 1093.0 Anthropic: Claude Opus 4.5 — $5/M, score 1259.0 Anthropic: Claude Haiku 4.5 — $1/M, score 1135.0 Qwen: Qwen3 Coder 30B A3B Instruct — $0.07/M, score 1100.0 Z.ai: GLM 4.5 Air — $0.13/M, score 1159.0 Qwen: Qwen3 235B A22B Thinking 2507 — $0.23/M, score 1065.0 Qwen: Qwen3 235B A22B Instruct 2507 — $0.22/M, score 1070.0 DeepSeek: R1 0528 — $0.5/M, score 1161.0 Qwen: Qwen3 30B A3B — $0.12/M, score 967.0 Meta: Llama 4 Scout — $0.1/M, score 762.0 Amazon: Nova Pro 1.0 — $0.8/M, score 808.0 GPT-5 Nano — $0.05/M, score 1114.0 Step 3.7 Flash — $0.185/M, score 1206.0 Muse Spark 1.3 Input price per million tokens (log scale) Elo

The stepped line is the efficient frontier: at each price, the best score available for that money or less. A model sitting on it is not being beaten by anything cheaper. Price is log-scaled because listed rates span four orders of magnitude. Only models with both a listed price and a score on this benchmark can appear.

Catalog activity

Change log

Field-level changes detected between successful source imports.

Full change log
Max Output Tokens943718 → 131072
Max Output Tokens131072 → 943718
Max Output Tokens943718 → 131072
Max Output Tokens131072 → 943718
Max Output Tokens943718 → 131072
Max Output Tokens131072 → 943718
Max Output Tokens943718 → 131072
Max Output Tokens131072 → 943718
Max Output Tokens943718 → 131072
Max Output Tokens131072 → 943718
Provenance

Sources & verification

Every figure on this page traces back to one of these records.

Methodology
Last verified
Sep 11, 2026
Status
Source-linked
Confidence
Medium
Catalog source
Models.dev
Models.devcatalog ↗Vercel AI Gateway documentationprovider documentation ↗Kilo Gateway documentationprovider documentation ↗OpenRouter documentationprovider documentation ↗NanoGPT documentationprovider documentation ↗DevPass (LLM Gateway) documentationprovider documentation ↗Meta documentationprovider documentation ↗EmpirioLabs AI documentationprovider documentation ↗OpenCode Zen documentationprovider documentation ↗Design Arena: websiteindependent benchmark ↗
Public API

Use this record

Fetch the complete source-linked model record. No key, no account, no rate-limited tier.

API documentation
Endpoint
GET https://model.kyssta.lol/api/v1/models/meta/muse-spark-1.3
curl
curl "https://model.kyssta.lol/api/v1/models/meta/muse-spark-1.3"
Common questions

Frequently asked questions

Answered directly from the stored record — nothing here is generated beyond the catalog's own fields.

What is Muse Spark 1.3?

Muse 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. It is published by Meta and catalogued here from Models.dev.

How much does Muse Spark 1.3 cost?

Listed input pricing starts at $1.25 per million tokens from Vercel AI Gateway.

What is the context length of Muse Spark 1.3?

Muse Spark 1.3 accepts up to 1.04858M tokens of context and returns up to 131.072K output tokens.

Does Muse Spark 1.3 support tool calling and structured output?

Provider catalogs list support for tool calling, structured output, reasoning, and image input.

Which providers serve Muse Spark 1.3?

9 providers list this model: Vercel AI Gateway, Kilo Gateway, OpenRouter, NanoGPT, LLM Gateway, DevPass (LLM Gateway) and 3 more.

Are the weights for Muse Spark 1.3 open?

No. This model is served through hosted APIs only.

When was Muse Spark 1.3 released?

The catalog records a release date of 2026-09-02, last verified Sep 11, 2026.

More models from Meta

View all →
Meta: Llama 3.2 1B Instruct

Llama 3.2 1B is a 1-billion-parameter language model focused on efficiently performing natural language tasks, such as summarization, dialogue, and multilingual text analysis. Its smaller size allows it to operate...

60K ctx $0.027/M in Meta: Llama 3.2 3B Instruct

Llama 3.2 3B is a 3-billion-parameter multilingual large language model, optimized for advanced natural language processing tasks like dialogue generation, reasoning, and summarization. Designed with the latest transformer architecture, it...

131.072K ctx $0.05/M in Meta: Llama 4 Maverick

Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward...

128K ctx $0.2/M in Meta: Llama 3.3 70B Instruct

The Meta Llama 3.3 multilingual large language model (LLM) is a pretrained and instruction tuned generative model in 70B (text in/text out). The Llama 3.3 instruction tuned text only model...

131.072K ctx $0.1/M in