Muse Ember vs Hy4 preview
Muse Ember leads Hy4 preview on coding, 64.9 to 44.3 on the Vibe Coding Index, and Muse Ember costs 5.0x less per output token ($0.50 vs $2.501 per million). Updated November 14, 2026; 9 benchmarks have a result for both.
Muse Ember or Hy4 preview: which to pick
Pick Muse Ember for coding: it scores 64.9 on the Vibe Coding Index against 44.3 for Hy4 preview.
Pick Hy4 preview for work measured by LMArena WebDev or LMArena WebDev: React, where it leads.
- Shared results
- 9
- Muse Ember only
- 6
- Hy4 preview only
- 16
- Like-for-like categories
- 3 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, six public coding benchmarks
Muse Ember
Muse Ember scores 64.9 against 44.3.
3 benchmarks behind each score
General
Vals Index
Muse Ember
Muse Ember leads, 60.5% against 53.2%.
Same board, both measured
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Muse Ember
Muse Ember $0.0003, Hy4 preview $0.0021 per turn.
List prices
At a glance
- Coding index
- Muse Ember 64.9 from 3 benchmarks, Hy4 preview 44.3 from 3 benchmarks.
- Widest category
- Coding, 10.2 points apart on average over 5 shared benchmarks.
- Widest single gap
- Terminal-Bench 4 (Vals AI): Muse Ember 39.4% against Hy4 preview 8.1%, 41.7 points apart on a 0 to 100 scale.
- Output price
- Muse Ember $0.50 against Hy4 preview $2.501 per million output tokens, 5.0x apart.
- Context window
- Both take 1.05M tokens.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | Muse Ember | Bars | Hy4 preview | Ties |
|---|---|---|---|---|
| Coding | 2 | 3 | ||
| Agentic work | 3 | 0 | ||
| General | 1 | 0 | ||
| All shared boards | 6 | 3 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index23
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| LMArena WebDev | 1609 | 1632 | Hy4 preview |
| Terminal-Bench 4 (Vals AI) | 39.4% | 8.1% | Muse Ember |
| Vibe Code Bench (Vals AI) | 83.7% | 77.5% | Muse Ember |
| Code migration (Vals AI) | Not measured | 47.4% | Only Hy4 preview measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only Hy4 preview measured |
| IOI (Vals AI) | Not measured | 59.3% | Only Hy4 preview measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 55.1% | Only Hy4 preview measured |
| CyberBench (Vals AI) | Not measured | 65.4% | Only Hy4 preview measured |
| Terminal-Bench 4.0 (Artificial Analysis run) | 40.9% | Not measured | Only Muse Ember measured |
| FrontierCode 2.0 | 27.2% | Not measured | Only Muse Ember measured |
| LMArena WebDev: React | 1616 | 1653 | Hy4 preview |
| LMArena WebDev: HTML | 1621 | 1625 | Hy4 preview |
| GameBench 2 | Not measured | 60.8 | Only Hy4 preview measured |
| Neither model has a result on the other 39 boards here | |||
IntelligenceNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| Neither model has a result on 2 boards here | |||
Agentic workLike-for-likePublic boards, shown as published, kept out of the index30
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| SREBench (Vals AI) | Not measured | 2.3% | Only Hy4 preview measured |
| Finance Agent (Vals AI) | Not measured | 55.1% | Only Hy4 preview measured |
| Tax Agent Bench (Vals AI) | 38.2% | 32.6% | Muse Ember |
| Legal Research Bench (Vals AI) | 49.0% | 45.2% | Muse Ember |
| Harvey's Legal Agent Benchmark (Vals AI) | 25.0% | 8.8% | Muse Ember |
| Excel Modeling Benchmark (Vals AI) | Not measured | 58.1% | Only Hy4 preview measured |
| BioMysteryBench (Vals AI) | Not measured | 69.3% | Only Hy4 preview measured |
| Terminal-Bench-Science (Vals AI) | Not measured | 1.4% | Only Hy4 preview measured |
| APEX-Agents | 64.3% | Not measured | Only Muse Ember measured |
| Neither model has a result on the other 35 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index2 boards, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| HLE-Diamond (Artificial Analysis run) | 25.1% | Not measured | Only Muse Ember measured |
| CritPt v2, hard subset (Artificial Analysis run) | 39.0% | Not measured | Only Muse Ember measured |
| Neither model has a result on the other 24 boards here | |||
MathDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| ProofBench | Not measured | 75.0% | Only Hy4 preview measured |
| Neither model has a result on the other 13 boards here | |||
KnowledgeDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| LegalBench (Vals AI) | Not measured | 83.8% | Only Hy4 preview measured |
| MedCode (Vals AI) | Not measured | 43.3% | Only Hy4 preview measured |
| AA-Omniscience | 7.7 | Not measured | Only Muse Ember measured |
| Neither model has a result on the other 18 boards here | |||
Web designNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| Neither model has a result on 9 boards here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| Vals Index | 60.5% | 53.2% | Muse Ember |
| Public Benefits Bench (Vals AI) | Not measured | 68.6% | Only Hy4 preview measured |
| MedScribe (Vals AI) | Not measured | 83.6% | Only Hy4 preview measured |
| Neither model has a result on the other 46 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Muse Ember | Hy4 preview | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
Muse Ember costs less
Repository review
50K input + 3K output tokens
Muse Ember costs less
Agent loop
200K input + 10K output tokens
Muse Ember costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | Meta Muse Ember | Tencent Hy4 preview |
|---|---|---|
| Maker | Meta | Tencent |
| Weights | No public weights listed | Open weights |
| Released | November 9, 2026 | August 28, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $0.10 | $0.834 |
| Output $/Mtok | $0.50 | $2.501 |
| Coding index | 64.9 | 44.3 |
| Independent measurements behind it | 3 | 3 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Muse Ember better than Hy4 preview for coding?
Pick Muse Ember for coding: it scores 64.9 on the Vibe Coding Index against 44.3 for Hy4 preview. Pick Hy4 preview for work measured by LMArena WebDev or LMArena WebDev: React, where it leads.
Which is cheaper, Muse Ember or Hy4 preview?
Muse Ember costs $0.10 per million input tokens and $0.50 per million output tokens; Hy4 preview costs $0.834 and $2.501. Cost per solved task is known for neither of them.
Which has the bigger context window, Muse Ember or Hy4 preview?
Neither: both take 1.05M tokens.
How many benchmarks is the Muse Ember vs Hy4 preview comparison based on?
9 benchmarks have a result for both models as of November 14, 2026. Muse Ember's coding index rests on 3 independent benchmarks and Hy4 preview's on 3. Where only one model has a result, the page says so instead of naming a winner.