Muse Spark Open 816B A42B
Muse Spark Open 816B A42BCompare inon
Muse Spark Open 816B A42B measured on 5 of 104 Chat & Knowledge benchmarks5
BenchmarkByScoreRank
Knowledge1
Math2
Legacy · saturated or no longer reported2
No benchmark matches
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperClaude Haiku 5.5xhighsame-90%$0.12Remove sampling parameters+5
OpenDeepSeek V4 Pro 0813maxsamesame$0.67Pass back reasoning_content · Chat Completions+4
CheaperGPT-6 Lunamaxsame-90%$0.068Function calling only with none · Chat Completions+3
CheaperGemini 3.8 Flashhighsame-25%*$1.24Sampling parameters are ignored+1CheaperGPT-5.6 Lunamaxsame-78%$0.18ModelScoreCostPer taskWhat changes
Claude Haiku 5.5xhighsame-90%$0.12Remove sampling parameters+5GPT-6 Lunamaxsame-90%$0.068Function calling only with none · Chat Completions+3Gemini 3.8 Flashhighsame-25%*$1.24Sampling parameters are ignored+1GPT-5.6 Lunamaxsame-78%$0.18
Kimi K2.6unstatedOpen-2%-14%$0.95 / $4ModelScoreCostPer taskWhat changes
Nothing here yet.
ModelScoreCostPer taskWhat changes
DeepSeek V4 Pro 0813maxOpensamesame$0.67Pass back reasoning_content · Chat Completions+4Kimi K3maxOpensame+200%$2.00Sampling parameters are fixed+3Kimi K2.6unstatedOpen-2%-14%$0.95 / $4Kimi K2.7 CodeunstatedOpen-3%-14%$0.95 / $4Instant mode is not supported · Self-hosted / vLLM / SGLang
GLM 5.3 FlashmaxOpen-4%-88%$0.28Defaults to max · Self-hosted+4ModelScoreCostPer taskWhat changes
Muse Wondermax+1%—$4 / $20Only auto tool_choice is supported+5Muse Spark 1.3xhigh+1%—$1.37Only auto tool_choice is supported+5Muse Sparkunstated-9%—$1.25 / $4.25Only n=1 · Meta / Chat Completions+1Llama 4 Maverick 17B 128E Instruct FP8Open-79%—$0.20 / $0.80Llama 3.1 405B Instruct-90%——Llama 4 ScoutunstatedOpen-92%—$0.10 / $0.30Llama 3.3 70B InstructunstatedOpen-95%—$0.10 / $0.32Meta Llama 3 70B InstructOpen-96%—$0.30 / $0.40All Meta models →Benchmarks
in Chat & KnowledgeGPQA Diamond · Epoch88.6#39
OTIS Mock AIME 2024 202598.1#23
FrontierMath Tiers 1–3 (v2)69.8#24
FrontierMath Tier 4 (v2)31.7#25
SimpleQA Verified v156.6#18
FAQ
How much does Muse Spark Open 816B A42B cost?›
$1.25 per 1M input tokens and $4.25 per 1M output tokens; cached input $0.15 per 1M (cheapest host: together).
Is Muse Spark Open 816B A42B free to use?›
There is no free API; the cheapest paid price is $1.25 per 1M input tokens.
How good is Muse Spark Open 816B A42B, and what is it best at?›
#18 of 83 in Chat & Knowledge.
Is Muse Spark Open 816B A42B better than Grok 4.7?›
On OTIS Mock AIME 2024 2025, Muse Spark Open 816B A42B scores 98.1 vs 98.1 — about the same.
How can I use Muse Spark Open 816B A42B?›
Through the Meta API.
What is Muse Spark Open 816B A42B?›
Muse Spark Open 816B A42B is a language model from Meta.
What are the best alternatives to Muse Spark Open 816B A42B?›
Claude Haiku 5.5 (cheaper), DeepSeek V4 Pro 0813 (open weights).