Muse Spark 1.2 API Benchmarks, Pricing & Provider Data
Compare Muse Spark 1.2 with another model
Choose a model to open its comparison page.
The page also shows measured API speed and first-token latency.
Muse Spark 1.2 is a reasoning model from Meta, designed for complex agentic tasks. It accepts text, images, video, audio, and PDF documents, returns text, and offers a 1M-token context...
Category Performance
Observed-capability estimates with uncertainty reported separately from independent benchmark families.
- Coverage
- 4 / 8
- Methodology
- V3.0
#1Knowledge58.8Provisional1/4 Measured dimensions
#2Reasoning58.5Estimated2/4 Measured dimensions
#3Agents56.3RatedGlobal rank #203/4 Measured dimensions
#4Coding55.7Estimated2/4 Measured dimensions
Specifications
Input and output token limits for this model, plus how it ranks on long-context understanding.
Features
Rankings
Excels at
Falls behind in
Detailed scores
Updated: Sep 14, 2026Overall
undefined metric} other undefined metrics}}
Overall score69.0#20 / 112DeepSWE59.3#14 / 21
Overall
undefined metric} other undefined metrics}}
Speed & latency
undefined metric} other undefined metrics}}
Output speed211.3 tok/s#17 / 80Time to first token12.79 s#68 / 80
Speed & latency
undefined metric} other undefined metrics}}
Pricing
undefined metric} other undefined metrics}}
Input price$1.25/M#123 / 186Output price$4.25/M#116 / 186
Pricing
undefined metric} other undefined metrics}}
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Score56.3#2080% interval47.4–65.33/4 Measured dimensionsTerminal-Bench 2.182.9#7 / 10GDPval-AA1631.0#9 / 72
Agents
V3.0undefined metric} other undefined metrics}} · Rated
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Score55.780% interval43.8–67.62/4 Measured dimensionsSciCode57.4%#9 / 89Terminal-Bench 2.182.9#7 / 11
Coding
V3.0undefined metric} other undefined metrics}} · Estimated
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Score58.580% interval47.7–69.22/4 Measured dimensionsGPQA90.4%#32 / 218HLE45.5%#17 / 216
Reasoning
V3.0undefined metric} other undefined metrics}} · Estimated
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Score58.880% interval44.8–72.81/4 Measured dimensionsKnowledge score76.0#26 / 83Artificial Analysis Intelligence Index39.8#25 / 117
Knowledge
V3.0undefined metric} other undefined metrics}} · Provisional
Math
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Math
V3.0undefined metric} other undefined metrics}} · No data
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Multilingual
V3.0undefined metric} other undefined metrics}} · No data
Multimodal
V3.0undefined metric} other undefined metrics}} · No data
80% interval30.8–69.20/4 Measured dimensionsDesign Arena Website1322.0#3 / 78
Multimodal
V3.0undefined metric} other undefined metrics}} · No data
Instruction following
V3.0undefined metric} other undefined metrics}} · No data
No data80% interval30.8–69.20/4 Measured dimensions
Instruction following
V3.0undefined metric} other undefined metrics}} · No data
OpenRouter endpoints
1 endpointsThird-party OpenRouter endpoint data, shown separately from LMSpeed measurements. Some 30-minute live performance fields only appear after syncing with an OpenRouter API key.
| Provider endpoint | Input | Output | 1d uptime | 30m latency | 30m throughput | Context / output |
|---|---|---|---|---|---|---|
Meta meta | $1.25/M | $4.25/M | 100% | — | — | undefined tokens / undefined tokens |
Alternatives & Similar Models
GPT-5.4
gpt-5-4
OpenAI GPT-5.4 extends the GPT-5 family with stronger instruction following, deeper tool use, and improved performance on coding, math, and long-document analysis.
GPT-5.3 Codex
gpt-5-3-codex
OpenAI GPT-5.3 Codex is a code-specialized variant in the GPT-5 series, optimized for code generation, debugging, and software development tasks.
GPT-5.4 Mini
gpt-5-4-mini
OpenAI GPT-5.4 Mini is a compact language model in the GPT-5 series, optimized for quick responses and high throughput.
DeepSeek V4 Flash
deepseek-v4-flash
DeepSeek V4 Flash is a fast, cost-efficient language model in the DeepSeek V4 family, optimized for low-latency chat, coding assistance, and high-throughput API workloads while retaining strong reasoning quality.
Claude Opus 4.6
claude-opus-4-6
Anthropic Claude Opus 4.6 is the most capable Claude Opus tier, optimized for complex analysis, long-horizon coding, and high-stakes enterprise reasoning workloads.
Claude Sonnet 4.6
claude-sonnet-4-6
Anthropic Claude Sonnet 4.6 extends the Sonnet line with improved tool use, coding reliability, and long-context performance for everyday production workloads.