Inkling Small vs Muse Ember
Full profiles: Inkling Small (Thinking Machines) and Muse Ember (Meta): every benchmark, price and rank.
Updated · Data from 9 sources
As of 16 Nov 2026, Inkling Small and Muse Ember are neck and neck on quality (52.0 vs 52.1 on the BenchLeader Index, inside the ±6.4 margin), and Muse Ember is 2.6× cheaper at $0.200 vs $0.525 per 1M tokens blended. Choose Inkling Small for knowledge (54 vs 49); choose Muse Ember for fast first answers (5.0 s vs 15.6 s) and coding (60 vs 46), reasoning (51 vs 43), and agents & tools (52 vs 46), where it also streams 3×+ faster.
Key facts (as of 16 Nov 2026)
- Inkling Small scores 52.0 on the BenchLeader Index and Muse Ember scores 52.1 (BenchLeader, 16 Nov 2026).
- Inkling Small costs $0.525 per 1M tokens blended (list price, 16 Nov 2026).
- Muse Ember costs $0.200 per 1M tokens blended (list price, 16 Nov 2026).
- Muse Ember streams 3×+ faster than Inkling Small (Artificial Analysis, 7-day median, 16 Nov 2026).
- Of the 24 benchmarks both models have, Muse Ember leads 23 (BenchLeader, 16 Nov 2026).
- On Arena WebDev, Inkling Small is rated 1408 and Muse Ember 1609 (LMArena, as published by 16 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Muse Ember | 60 vs 46 category score; Arena WebDev 1609 vs 1408 | High |
| Agents & tool use | Muse Ember | 52 vs 46 category score; Arena Agent 1.7 vs -18.1 | Medium |
| Reasoning | Muse Ember | 51 vs 43 category score; HLE-Diamond (AA) 25.1% vs 12.1% | High |
| Knowledge | Inkling Small | 54 vs 49 category score; AA-Omniscience v1.1 -9.5 vs 7.4 | Medium |
| Chat quality (human preference) | – | Only Inkling Small has an Arena rating (58 category score) | Not comparable |
| Lowest cost | Muse Ember | $0.200 vs $0.525 per 1M blended; agentic coding session $0.008 vs $0.023 | – |
| Fastest first answer | Muse Ember | 5.0 s (Muse Ember (max)) vs 15.6 s (Inkling Small) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Inkling Small and Muse Ember (max) are neck and neck on quality (52.0 vs 52.1, within the ±6.4 margin).
- Inkling Small is stronger in knowledge.
- Muse Ember (max) is stronger in agents & tools, coding, reasoning.
- Muse Ember (max) is 2.6× cheaper ($0.200 vs $0.525 per 1M blended).
- Muse Ember (max) streams 3.8× faster (437 vs 116 tokens per second).
At a glance
| Spec | Inkling Small | Muse Ember |
|---|---|---|
| Developer | Thinking Machines | Meta |
| Released | 15 Jul 2026 | 9 Nov 2026 |
| Context window (tokens) | 65,536 | 1,048,576 |
| Compared at | Inkling Small | Muse Ember (max) |
| Open weights | Yes | No |
| Price in / out / cached input per 1M | $0.300 / $1.20 / $0.116 | $0.100 / $0.500 / $0.0040 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Inkling Small | Muse Ember (max) | Leader |
|---|---|---|---|
| Reasoning | 43.2 | 51.1 | Muse Ember |
| Coding | 45.6 | 60.0 | Muse Ember |
| Agents & tools | 45.7 | 52.4 | Muse Ember |
| Maths | 53.0 | – | not comparable |
| Knowledge | 54.3 | 48.8 | Inkling Small |
| Human preference | 57.8 | – | not comparable |
| Multimodal | 56.3 | – | not comparable |
| Long context | 62.4 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Maths, Instruction following, Long context, Multimodal, Human preference.
Where sources disagree
Across the 24 benchmarks both have, Inkling Small leads 1, Muse Ember leads 23. Every overall index that rates both puts Muse Ember ahead.
| Overall index | Inkling Small | Muse Ember (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 52.0 | 52.1 | neck and neck |
| AA Intelligence Index v5.0.1 | 12.9 | 36.0 | Muse Ember |
| Vals Index | 27.6 | 60.5 | Muse Ember |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Inkling Small | Muse Ember (max) |
|---|---|---|
| Blended price per 1M tokens | $0.525 | $0.200 |
| Input / output per 1M | $0.300 / $1.20 | $0.100 / $0.500 |
| Cached input per 1M | $0.116 | $0.0040 |
| Price per task: chat reply | $0.0005 | $0.0002 |
| Price per task: summarize a 30-page report | $0.0043 | $0.0015 |
| Price per task: code edit | $0.0036 | $0.0014 |
| Price per task: agentic coding session | $0.023 | $0.0080 |
| Price per task: structured extraction | $0.0008 | $0.0003 |
| Agentic coding session, input mostly cached | $0.015 | $0.0037 |
| Cost per benchmark task (AA Index) | $0.203 | $0.136 |
| Output tokens per task (AA Index) | 37k | 84k |
| Of which reasoning | 25k | 52k |
| Time per task (AA Index) | 5 min | 3 min |
| Output speed | 116 tok/s | 437 tok/s |
| Time to first answer | 15.6 s | 5.0 s |
| Time to a complete answer, last measured 13 Nov 2026 | – | 12.9 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $0.203 on Inkling Small and $0.136 on Muse Ember (max), and Muse Ember (max) wrote 2.3× as many output tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. Inkling Small scores 52.0 and takes 16 s to its first answer token; Inkling Small (xhigh) scores 47.3 in 16 s. Muse Ember (max) scores 52.1 and takes 5 s to its first answer token; Muse Ember (xhigh) scores 49.1 in 4 s. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
101 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Inkling Small | Muse Ember (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 52.0 | 52.1 |
| Agents & tools score | BenchLeader | 45.7 | 52.4 |
| Coding score | BenchLeader | 45.6 | 60.0 |
| Human preference score | BenchLeader | 57.8 | – |
| Knowledge score | BenchLeader | 54.3 | 48.8 |
| Long context score | BenchLeader | 62.4 | – |
| Maths score | BenchLeader | 53.0 | – |
| Multimodal score | BenchLeader | 56.3 | – |
| Reasoning score | BenchLeader | 43.2 | 51.1 |
| Blended price $/M | BenchLeader | $0.525 | $0.200 |
| Output speed | BenchLeader | 116 tok/s | 437 tok/s |
| Time to first answer | BenchLeader | 15.6 s | 5.0 s |
| Context window | BenchLeader | 65,536 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 88.5%† | – |
| SimpleBench | SimpleBench | – | 72.7%† |
| Arena Hard Prompts | Arena (LMArena) | 1428 | – |
| GPQA Diamond (AA) | Artificial Analysis | 89.5% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 33.3% | – |
| HLE-Diamond (AA) | Artificial Analysis | 12.1% | 25.1% |
| GPQA Diamond (Vals) | Vals AI | 83.6% | – |
| ARC-AGI-1 | ARC Prize | 84.0%† | – |
| ARC-AGI-2 | ARC Prize | 40.1%† | – |
| CritPt | Artificial Analysis | 8.3% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 18.5% | 39.0% |
| Arena Expert | Arena (LMArena) | 1441 | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 6.0%† | – |
| Coding | |||
| SciCode | SciCode | 49.6% | – |
| Arena Coding | Arena (LMArena) | 1472 | – |
| Arena WebDev | Arena (LMArena) | 1408 | 1609 |
| LiveCodeBench | Vals AI | 85.9% | – |
| IOI | Vals AI | 9.3% | – |
| SWE-bench (Vals) | Vals AI | 82.2% | – |
| Code Migration | Vals AI | 13.7% | – |
| Code Migration v1.1 | Vals AI | 10.4% | – |
| ProgramBench | Vals AI | 0.5% | – |
| Vibe Code Bench v1.1 | Vals AI | 19.1% | 83.7% |
| SWE Atlas v1.1 | Scale AI SEAL | – | 56.0%† |
| SWE Atlas v1.1: Codebase QnA | Scale AI SEAL | – | 61.3%† |
Show all 101 benchmarks
| Metric | Publisher | Inkling Small | Muse Ember (max) |
|---|---|---|---|
| Coding | |||
| SWE Atlas v1.1: Test Writing | Scale AI SEAL | – | 56.7%† |
| SWE Atlas v1.1: Refactoring | Scale AI SEAL | – | 50.0%† |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 1.0% | 40.9% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 55.1% | – |
| GameBench 2 Overall | DuelLab GameBench 2 | 29.2 | – |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 28.1† | – |
| Design Arena Website | Design Arena | – | 1330† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 0.6% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 4.8% | 27.2% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 0.9%† | 30.6%† |
| APEX-Agents | Mercor | – | 64.3%† |
| Arena Agent | Arena (LMArena) | -18.1 | 1.7 |
| Terminal-Bench 2.1 (Vals) | Vals AI | 55.1% | – |
| MCP Atlas | Scale AI SEAL | 79.2% | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 31.2% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1133 | – |
| τ³-Banking (AA) | Artificial Analysis | 18.8% | – |
| Analyst Agent (AA) | Artificial Analysis | 27.5% | – |
| CyberBench | Vals AI | 64.1% | – |
| Finance Agent v2 | Vals AI | 41.3% | – |
| Finance Agent v2.1 | Vals AI | 44.8% | 63.8% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 2.9% | 25.0% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 1.7% | – |
| Legal Research Bench | Vals AI | 25.5% | 49.0% |
| SkillsBench | Vals AI | 33.6% | – |
| Tax Agent Bench | Vals AI | 15.4% | 38.2% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 0.0% | 39.4% |
| Arena Agent Task Completion | Arena (LMArena) | -27.6 | -7.6 |
| AutomationBench | Artificial Analysis | 4.8% | 68.1% |
| GDP.pdf (AA) | Artificial Analysis | 11.2% | 22.8% |
| Harvey LAB-AA v1.1 | Harvey | 0.0% | – |
| EnterpriseOps-Gym | Artificial Analysis | 42.7% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 912 | 1520 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 0.0% | 20.0% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 46.3%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 17.1%† | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 90.0%† | – |
| ProofBench | Vals AI | 6.0% | – |
| Arena Maths | Arena (LMArena) | 1452 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 19.1%† | – |
| AA-Omniscience v1.1 | Artificial Analysis | -9.5 | 7.4 |
| MMLU-Pro | Vals AI | 85.6% | – |
| LegalBench | Vals AI | 83.0% | – |
| CorpFin | Vals AI | 69.6% | – |
| TaxEval | Vals AI | 75.5% | – |
| Excel Modeling Benchmark | Vals AI | 31.8% | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 40.8% | 72.2% |
| MedCode | Vals AI | 37.9% | – |
| MedScribe | Vals AI | 84.1% | – |
| Public Benefits Bench v1.1 | Vals AI | 59.4% | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 31.8% | 21.1% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 39.5% | 82.6% |
| Instruction following | |||
| Arena Instruction Following | Arena (LMArena) | 1397 | – |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1405 | – |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1491 | – |
| Arena Creative Writing | Arena (LMArena) | 1319 | – |
| Arena Multi-turn | Arena (LMArena) | 1414 | – |
| Arena Longer Queries | Arena (LMArena) | 1401 | – |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1236 | – |
| MMMU-Pro | Artificial Analysis | 74.0% | – |
| MortgageTax | Vals AI | 62.3% | – |
| AudioMC | Scale AI SEAL | 54.9% | – |
| AudioMC (text output) | Scale AI SEAL | 54.9% | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 75.7% | – |
| MLCR | Artificial Analysis | 7.8% | 18.3% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 150.8 | – |
| AA Intelligence Index v5.0.1 | Artificial Analysis | 12.9 | 36.0 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 25.7 | – |
| AA Progress Index v1 | Artificial Analysis | 282.7 | – |
| Vals Index | Vals AI | 27.6 | 60.5 |
| Vals Finance Index | Vals AI | 36.3% | 60.9% |
| Vals Multimodal Index | Vals AI | 50.0% | – |
| Vals Router Index | Vals AI | 27.6% | – |
| Vals Legal Index | Vals AI | 57.8% | 75.7% |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | – | 11.9%† |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 33.3 | – |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261111-7b598bb13de4 (https://benchmarks.duellab.org/releases/gb2-20261111-7b598bb13de4/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- Inkling Small: 1 score revision.
- Muse Ember (max): 1 new result (Arena Agent).
- Inkling Small: 1 score revision.
- Muse Ember (max): 2 new results (MLCR, Arena WebDev).
- Inkling Small: 1 score revision.
- Inkling Small: 1 score revision.
- Inkling Small: 1 score revision.
- Inkling Small: BenchLeader Index 54.9 → 51.9.
- Inkling Small: 5 new results (AA Intelligence Index v5.0.1, AA Progress Index v1, HLE-Diamond (AA)…); 3 score revisions; Inkling Small cost per task moved from $0.089/M to $0.20/M (127% up).
- Inkling Small: 1 new result (AA-Briefcase v1.1).
Related comparisons
- Inkling Small vs Inkling 2 Small
- Inkling Small vs MiniMax M3.1 Flash
- Muse Ember vs Muse Spark Open 816B A42B
- Muse Ember vs Mistral Large 4
- Muse Ember vs Gemini 3.1 Flash Lite
- Muse Ember vs Qwen3.5-27B
- Muse Ember vs Gemma 4 31B
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Inkling Small vs Muse Ember: questions
- Is Inkling Small better than Muse Ember?
- Inkling Small and Muse Ember (max) are neck and neck on quality (52.0 vs 52.1, within the ±6.4 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 16 Nov 2026, so check the category scores for your use.
- Inkling Small vs Muse Ember: which is cheaper, and will that change?
- Muse Ember is cheaper today: $0.200 against $0.525 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Inkling Small or Muse Ember better for coding?
- Muse Ember scores higher in coding (60 vs 46 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1408 for Inkling Small against 1609 for Muse Ember.
- Which is faster, Inkling Small or Muse Ember?
- Muse Ember streams 3×+ faster (437 vs 116 output tokens per second, Artificial Analysis 7-day median).
- Is Inkling Small or Muse Ember better for agents and tool use?
- Muse Ember scores higher in agentic tasks (52 vs 46 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is -18.1 for Inkling Small against 1.7 for Muse Ember.
More questions (4)
- Which is better for long documents?
- Muse Ember accepts more context: 1,048,576 against 65,536 tokens for Inkling Small.
- Which has lower latency?
- Muse Ember answers first: 5.0 s against 15.6 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on Inkling Small vs Muse Ember?
- Across the 24 benchmarks both have, Inkling Small leads 1, Muse Ember leads 23. Every overall index that rates both puts Muse Ember ahead.
- When was Muse Ember released?
- Muse Ember was released on 9 Nov 2026.