Meta's Muse Ember matches Claude Haiku 5.5 on Intelligence Index at a third of the cost
ArtificialAnlys · x · 2026-11-10
Artificial Analysis evaluated Meta's new efficient-tier model line Muse Ember: it scores 36 on the Intelligence Index, matching Claude Haiku 5.5 (max) at $0.14 per task — roughly a third of the cost.
Key results
- 36 at max effort ($0.14/task), 34 at xhigh ($0.12); sits on the intelligence-vs-cost Pareto frontier, 4x cheaper than DeepSeek V4.1 Flash 1017 (33, $0.60) and GLM-5.3-Flash (28, $0.58)
- 8 points behind Meta's flagship Muse Wonder (44) at about 1/50th of the cost per task
- AutomationBench-AA 68.1%, level with Muse Wonder; beats Haiku 5.5 on GDP.pdf (22.8% vs 20.8%) but trails on AA-Briefcase (1521 vs 1574 Elo)
- Mixed agentic coding: wins Terminal-Bench 4.0 (40.9% vs 32.8%), loses FrontierCode 2.0 (27.2% vs 34.9%); scores 46 on the Coding Agent Index inside its own Muse Code, ahead of Haiku in Claude Code (37) and GPT-6 Luna in Codex (31)
Knowledge weak, hallucination rare: AA-Omniscience accuracy is just 21.1%, the lowest among comparable models, mostly because it declines 64% of questions — keeping hallucination rate at 17.4%, half of Haiku 5.5's.
Pricing & availability: $0.10 input / $0.004 cached input / $0.50 output per 1M tokens, via Meta's first-party API and Muse Code.
More from Infra
- Gary Marcus slams Nvidia after report says Commerce inquiry was dropped — GaryMarcus · 2026-11-10
- Claude Fable 5.2 and Mythos 5.2 spotted on AWS Bedrock, priced 20% below 5.1 — testingcatalog · 2026-11-10
- e-means paper trick: IVF approximate assignment cuts clustering cost 34x at 10B vectors — antoine_chaffin · 2026-11-10
- Telekom, NVIDIA and SAP expand German AI cloud, boosting capacity 50% — NinaDSchick · 2026-11-10
- Two-person YC-backed Archgen wins chip layout contest against 131 entries — testingcatalog · 2026-11-10
- NVIDIA Dynamo in 5 minutes: OpenAI-compatible frontend enables backend hot-swap and KV cache reuse — NVIDIA Developer · 2026-11-10