Mistral Large 4 vs Muse Ember
Full profiles: Mistral Large 4 (Mistral AI) and Muse Ember (Meta): every benchmark, price and rank.
Updated · Data from 8 sources
As of 16 Nov 2026, Mistral Large 4 and Muse Ember are neck and neck on quality (52.9 vs 52.1 on the BenchLeader Index, inside the ±6.0 margin), and Muse Ember is 5.3× cheaper at $0.200 vs $1.05 per 1M tokens blended. Muse Ember leads in every category we measure both in. Muse Ember streams 3×+ faster.
Key facts (as of 16 Nov 2026)
- Mistral Large 4 scores 52.9 on the BenchLeader Index and Muse Ember scores 52.1 (BenchLeader, 16 Nov 2026).
- Mistral Large 4 costs $1.05 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 Mistral Large 4 (Artificial Analysis, 7-day median, 16 Nov 2026).
- Of the 15 benchmarks both models have, Muse Ember leads 13 (BenchLeader, 16 Nov 2026).
- On Arena WebDev, Mistral Large 4 is rated 1565 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 54 category score; Arena WebDev 1609 vs 1565 | Medium |
| Agents & tool use | Muse Ember | 52 vs 49 category score; Arena Agent 1.7 vs -5.4 | Medium |
| Reasoning | Muse Ember | 51 vs 46 category score; HLE-Diamond (AA) 25.1% vs 15.1% | Medium |
| Knowledge | Muse Ember | 49 vs 41 category score; AA-Omniscience v1.1 7.4 vs -7.5 | High |
| Chat quality (human preference) | – | Only Mistral Large 4 has an Arena rating (60 category score) | Not comparable |
| Lowest cost | Muse Ember | $0.200 vs $1.05 per 1M blended; agentic coding session $0.008 vs $0.050 | – |
| Fastest first answer | Muse Ember | 5.0 s (Muse Ember (max)) vs 40.9 s (Mistral Large 4) 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
- Mistral Large 4 and Muse Ember (max) are neck and neck on quality (52.9 vs 52.1, within the ±6.0 margin).
- Muse Ember (max) is stronger in agents & tools, coding, knowledge, reasoning.
- Muse Ember (max) is 5.3× cheaper ($0.200 vs $1.05 per 1M blended).
- Muse Ember (max) streams 4.5× faster (437 vs 97 tokens per second).
At a glance
| Spec | Mistral Large 4 | Muse Ember |
|---|---|---|
| Developer | Mistral AI | Meta |
| Released | 6 Oct 2026 | 9 Nov 2026 |
| Context window (tokens) | 1,048,576 | 1,048,576 |
| Compared at | Mistral Large 4 | Muse Ember (max) |
| Open weights | Yes | No |
| Price in / out / cached input per 1M | $0.700 / $2.10 / $0.070 | $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 | Mistral Large 4 | Muse Ember (max) | Leader |
|---|---|---|---|
| Reasoning | 46.2 | 51.1 | Muse Ember |
| Coding | 53.8 | 60.0 | Muse Ember |
| Agents & tools | 48.8 | 52.4 | Muse Ember |
| Maths | 54.2 | – | not comparable |
| Knowledge | 40.7 | 48.8 | Muse Ember |
| Human preference | 59.7 | – | not comparable |
| Multimodal | 58.0 | – | not comparable |
| Long context | 65.3 | – | 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 15 benchmarks both have, Mistral Large 4 leads 2, Muse Ember leads 13. Every overall index that rates both puts Muse Ember ahead.
| Overall index | Mistral Large 4 | Muse Ember (max) | Leader |
|---|---|---|---|
| BenchLeader Index | 52.9 | 52.1 | neck and neck |
| AA Intelligence Index v5.0.1 | 25.3 | 36.0 | 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 | Mistral Large 4 | Muse Ember (max) |
|---|---|---|
| Blended price per 1M tokens | $1.05 | $0.200 |
| Input / output per 1M | $0.700 / $2.10 | $0.100 / $0.500 |
| Cached input per 1M | $0.070 | $0.0040 |
| Price per task: chat reply | $0.0009 | $0.0002 |
| Price per task: summarize a 30-page report | $0.0097 | $0.0015 |
| Price per task: code edit | $0.0074 | $0.0014 |
| Price per task: agentic coding session | $0.050 | $0.0080 |
| Price per task: structured extraction | $0.0018 | $0.0003 |
| Agentic coding session, input mostly cached | $0.022 | $0.0037 |
| Cost per benchmark task (AA Index) | $4.21 | $0.136 |
| Output tokens per task (AA Index) | 135k | 84k |
| Of which reasoning | 102k | 52k |
| Time per task (AA Index) | 23 min | 3 min |
| Output speed | 97 tok/s | 437 tok/s |
| Time to first answer | 40.9 s | 5.0 s |
| Time to a complete answer, last measured 13 Nov 2026 | 119.0 s | 12.9 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $4.21 on Mistral Large 4 and $0.136 on Muse Ember (max), and Mistral Large 4 wrote 1.6× 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. Mistral Large 4 scores 52.9 and takes 41 s to its first answer token; Mistral Large 4 (high) scores 48.3 in 45 s, Mistral Large 4 (medium) scores 45.9 in 27 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
112 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 | Mistral Large 4 | Muse Ember (max) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 52.9 | 52.1 |
| Agents & tools score | BenchLeader | 48.8 | 52.4 |
| Coding score | BenchLeader | 53.8 | 60.0 |
| Human preference score | BenchLeader | 59.7 | – |
| Knowledge score | BenchLeader | 40.7 | 48.8 |
| Long context score | BenchLeader | 65.3 | – |
| Maths score | BenchLeader | 54.2 | – |
| Multimodal score | BenchLeader | 58.0 | – |
| Reasoning score | BenchLeader | 46.2 | 51.1 |
| Blended price $/M | BenchLeader | $1.05 | $0.200 |
| Output speed | BenchLeader | 97 tok/s | 437 tok/s |
| Time to first answer | BenchLeader | 40.9 s | 5.0 s |
| Context window | BenchLeader | 1,048,576 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 88.1%† | – |
| SimpleBench | SimpleBench | – | 72.7%† |
| Arena Hard Prompts | Arena (LMArena) | 1453 | – |
| LiveBench 2.0 Reasoning | LiveBench | 85.2%† | – |
| LiveBench 1.0 Reasoning | LiveBench | 83.9%† | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 35.0% | – |
| HLE-Diamond (AA) | Artificial Analysis | 15.1% | 25.1% |
| CritPt | Artificial Analysis | 10.6% | – |
| CritPt v2 hard subset (AA) | Artificial Analysis | 24.5% | 39.0% |
| MysteryMechanism | Vals AI | 12.2%† | – |
| Arena Expert | Arena (LMArena) | 1464 | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 33.0%† | – |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 17.0%† | – |
| LMCA | Epoch AI Benchmarking Hub | 35.8% | – |
| DTBench | Epoch AI Benchmarking Hub | 88.2% | – |
| Coding | |||
| SciCode | SciCode | 54.2% | – |
| FrontierCode | Cognition | 39.4%† | – |
| Arena Coding | Arena (LMArena) | 1484 | – |
| Arena WebDev | Arena (LMArena) | 1565 | 1609 |
| LiveBench 2.0 Coding | LiveBench | 79.4%† | – |
| LiveBench 1.0 Coding | LiveBench | 77.2%† | – |
| IOI | Vals AI | 45.3%† | – |
| Code Migration | Vals AI | 30.6%† | – |
| SWE-bench Multimodal | Vals AI | 33.1%† | – |
| Vibe Code Bench 1-100 | Vals AI | 14.3%† | – |
Show all 112 benchmarks
| Metric | Publisher | Mistral Large 4 | Muse Ember (max) |
|---|---|---|---|
| Coding | |||
| Vibe Code Bench v1.1 | Vals AI | 78.4%† | 83.7% |
| SWE Atlas v1.1 | Scale AI SEAL | 45.2% | 56.0%† |
| SWE Atlas v1.1: Codebase QnA | Scale AI SEAL | 61.3% | 61.3%† |
| SWE Atlas v1.1: Test Writing | Scale AI SEAL | 40.0% | 56.7%† |
| SWE Atlas v1.1: Refactoring | Scale AI SEAL | 34.3% | 50.0%† |
| MirrorCode | Epoch AI Benchmarking Hub | 40.0%† | – |
| Surface Evolver Bench | Epoch AI Benchmarking Hub | 28.1% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1213.2 | – |
| WeirdML v3 | WeirdML | 4.5%† | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 26.8% | 40.9% |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 57.6† | – |
| Design Arena Website | Design Arena | 1258 | 1330† |
| FrontierCode 1.1 (AA) | Artificial Analysis | 41.5% | – |
| FrontierCode 2.0 (AA) | Artificial Analysis | 13.6% | 27.2% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 27.6%† | 30.6%† |
| APEX-Agents | Mercor | 51.6%† | 64.3%† |
| Arena Agent | Arena (LMArena) | -5.4 | 1.7 |
| LiveBench Long Agentic | LiveBench | 24.7%† | – |
| LiveBench Browser | LiveBench | 17.4%† | – |
| LiveBench Agentic Coding | LiveBench | 57.2%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 46.1% | – |
| GDPval-AA v2.1 | Artificial Analysis | 1431 | – |
| Finance Agent v2 | Vals AI | 54.7%† | – |
| Finance Agent v2.1 | Vals AI | 61.4%† | 63.8% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | 17.5%† | 25.0% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 15.8%† | – |
| Legal Research Bench | Vals AI | 31.7%† | 49.0% |
| Tax Agent Bench | Vals AI | 30.4%† | 38.2% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 27.8%† | 39.4% |
| Arena Agent Task Completion | Arena (LMArena) | -4.7 | -7.6 |
| AutomationBench | Artificial Analysis | 59.9% | 68.1% |
| GDP.pdf (AA) | Artificial Analysis | 18.6% | 22.8% |
| Harvey LAB-AA v1.1 | Harvey | 3.1% | – |
| AA-Briefcase v1.1 | Artificial Analysis | 1393 | 1520 |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 4.8% | 20.0% |
| CWE-Bench (AA) | Artificial Analysis | 50.8% | – |
| DeepSec-Bench (AA) | Artificial Analysis | 16.0% | – |
| CyberGym end-to-end (AA) | Artificial Analysis | 81.7% | – |
| MKA (AA) | Artificial Analysis | 35.1%† | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 58.6%† | – |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 26.8%† | – |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 98.3%† | – |
| ProofBench | Vals AI | 10.0% | – |
| LiveBench Mathematics | LiveBench | 93.6%† | – |
| Arena Maths | Arena (LMArena) | 1459 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 20.0%† | – |
| LiveBench 2.0 Data Analysis | LiveBench | 76.6%† | – |
| LiveBench Research | LiveBench | 80.3%† | – |
| LiveBench 1.0 Data Analysis | LiveBench | 76.5%† | – |
| AA-Omniscience v1.1 | Artificial Analysis | -7.5 | 7.4 |
| BioMysteryBench | Vals AI | 67.0%† | – |
| Excel Modeling Benchmark | Vals AI | 56.2%† | – |
| Excel Modeling Benchmark v1.1 | Vals AI | 69.0%† | 72.2% |
| MedCode | Vals AI | 40.7%† | – |
| MedScribe | Vals AI | 80.4%† | – |
| Immigration Bench | Vals AI | 63.7%† | – |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 23.9% | 21.1% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 58.7% | 82.6% |
| Instruction following | |||
| LiveBench Language | LiveBench | 49.6%† | – |
| LiveBench Instruction Following | LiveBench | 64.8%† | – |
| Arena Instruction Following | Arena (LMArena) | 1418 | – |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1420 | – |
| Arena Creative Writing | Arena (LMArena) | 1359 | – |
| Arena Multi-turn | Arena (LMArena) | 1429 | – |
| Arena Longer Queries | Arena (LMArena) | 1420 | – |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1244 | – |
| LiveBench Design | LiveBench | 76.3%† | – |
| MMMU-Pro | Artificial Analysis | 76.4% | – |
| Furniture Assembly | Epoch AI Benchmarking Hub | 20.8%† | – |
| VIALS-AA | Handshake AI | 20.5%† | – |
| Code2Video-AA | Artificial Analysis | 931 | – |
| AA-3D | Artificial Analysis | 882 | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 81.3% | – |
| MLCR | Artificial Analysis | 7.8% | 18.3% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 154.6 | – |
| LiveBench 2.0 | LiveBench | 60.3%† | – |
| LiveBench 1.0 | LiveBench | 71.8%† | – |
| AA Intelligence Index v5.0.1 | Artificial Analysis | 25.3 | 36.0 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 38.4 | – |
| AA Progress Index v1 | Artificial Analysis | 312.4 | – |
| Vals Index | Vals AI | – | 60.5 |
| Vals Finance Index | Vals AI | 56.9%† | 60.9% |
| Vals Legal Index | Vals AI | 65.6%† | 75.7% |
| AA Cyber Index | Artificial Analysis | 49.5 | – |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 23.1% | 11.9%† |
| DistressBench | Scale AI SEAL | 77.0% | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 38.9† | – |
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
- Mistral Large 4: BenchLeader Index 52.2 → 52.9.
- Mistral Large 4: 1 new result (GDPval-AA v2.1).
- Mistral Large 4: 1 score revision.
- Mistral Large 4: 2 score revisions.
- Muse Ember (max): 1 new result (Arena Agent).
- Mistral Large 4: 6 score revisions.
- Muse Ember (max): 2 new results (MLCR, Arena WebDev).
- Mistral Large 4: BenchLeader Index 52.7 → 52.2.
- Mistral Large 4: 7 score revisions.
- Mistral Large 4: 8 score revisions.
Related comparisons
- Mistral Large 4 vs Muse Spark 1.3
- Mistral Large 4 vs MiniMax M3.1 Flash
- Mistral Large 4 vs Mistral Large 4.0 OSS
- Muse Ember vs Muse Spark Open 816B A42B
- Muse Ember vs Gemini 3.1 Flash Lite
- Muse Ember vs Inkling Small
- Muse Ember vs Qwen3.5-27B
- Muse Ember vs Gemma 4 31B
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Mistral Large 4 vs Muse Ember: questions
- Is Mistral Large 4 better than Muse Ember?
- Mistral Large 4 and Muse Ember (max) are neck and neck on quality (52.9 vs 52.1, within the ±6.0 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.
- Mistral Large 4 vs Muse Ember: which is cheaper, and will that change?
- Muse Ember is cheaper today: $0.200 against $1.05 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Mistral Large 4 or Muse Ember better for coding?
- Muse Ember scores higher in coding (60 vs 54 on the category index, where 50 is average). On Arena WebDev, the top shared coding benchmark, it is 1565 for Mistral Large 4 against 1609 for Muse Ember.
- Which is faster, Mistral Large 4 or Muse Ember?
- Muse Ember streams 3×+ faster (437 vs 97 output tokens per second, Artificial Analysis 7-day median).
- Is Mistral Large 4 or Muse Ember better for agents and tool use?
- Muse Ember scores higher in agentic tasks (52 vs 49 on the category index, where 50 is average). On Arena Agent, the top shared agents & tools benchmark, it is -5.4 for Mistral Large 4 against 1.7 for Muse Ember.
More questions (4)
- Which is better for long documents?
- Both accept 1,048,576 tokens of context.
- Which has lower latency?
- Muse Ember answers first: 5.0 s against 40.9 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 Mistral Large 4 vs Muse Ember?
- Across the 15 benchmarks both have, Mistral Large 4 leads 2, Muse Ember leads 13. Every overall index that rates both puts Muse Ember ahead.
- When was Muse Ember released?
- Muse Ember was released on 9 Nov 2026.