Ember-1 vs Muse Spark Open
Ember-1 leads Muse Spark Open on coding, 52.6 to 47.4 on the Vibe Coding Index, and Muse Spark Open costs 3.5x less per output token ($4.25 vs $15.00 per million). Updated October 21, 2026; 7 benchmarks have a result for both.
Ember-1 or Muse Spark Open: which to pick
Pick Ember-1 for coding: it scores 52.6 on the Vibe Coding Index against 47.4 for Muse Spark Open, though on 2 benchmarks against 4.
Pick Muse Spark Open for output tokens 3.5x cheaper ($4.25 vs $15.00 per million) and work measured by Finance Agent (Vals AI), where it leads.
- Shared results
- 7
- Ember-1 only
- 7
- Muse Spark Open only
- 21
- Like-for-like categories
- 2 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
Ember-1
Ember-1 scores 52.6 against 47.4.
2 and 4 benchmarks behind the scores
Web design
Design Arena: websites
No call
Only Muse Spark Open has a Design Arena: websites result, 1317.
Directional only
General
Vals Index
No call
Only Muse Spark Open has a Vals Index result, 48.6%.
Directional only
Long documents
Listed context window
No call
Both take 1.05M tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
Muse Spark Open
Ember-1 $0.011, Muse Spark Open $0.0034 per turn.
List prices
At a glance
- Coding index
- Ember-1 52.6 from 2 benchmarks, Muse Spark Open 47.4 from 4 benchmarks.
- Widest category
- Coding, 12.6 points apart on average over 3 shared benchmarks.
- Widest single gap
- Terminal-Bench 4 (Vals AI): Ember-1 20.7% against Muse Spark Open 6.6%, 18.9 points apart on a 0 to 100 scale.
- Output price
- Ember-1 $15.00 against Muse Spark Open $4.25 per million output tokens, 3.5x 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 | Ember-1 | Bars | Muse Spark Open | Ties |
|---|---|---|---|---|
| Coding | 3 | 0 | ||
| Agentic work | 3 | 1 | ||
| All shared boards | 6 | 1 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index30
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| LiveBench coding | Not measured | 73.6% | Only Muse Spark Open measured |
| LMArena WebDev | Not measured | 1528 | Only Muse Spark Open measured |
| Terminal-Bench 4 (Vals AI) | 20.7% | 6.6% | Ember-1 |
| Vibe Code Bench (Vals AI) | 83.9% | 76.2% | Ember-1 |
| Code migration (Vals AI) | 32.5% | 21.2% | Ember-1 |
| IOI (Vals AI) | 46.3% | Not measured | Only Ember-1 measured |
| CyberBench (Vals AI) | 69.5% | Not measured | Only Ember-1 measured |
| SciCode | Not measured | 58.4% | Only Muse Spark Open measured |
| ALE-Bench | Not measured | 1590.1 | Only Muse Spark Open measured |
| LMArena WebDev: React | Not measured | 1529 | Only Muse Spark Open measured |
| LMArena WebDev: HTML | Not measured | 1495 | Only Muse Spark Open measured |
| Neither model has a result on the other 28 boards here | |||
IntelligenceDirectional onlyPublic boards, shown as published, kept out of the index1 board, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| Epoch Capabilities Index | Not measured | 155.1 | Only Muse Spark Open measured |
| Neither model has a result on the other 1 board here | |||
Agentic workLike-for-likePublic boards, shown as published, kept out of the index31
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| LiveBench research | Not measured | 80.3% | Only Muse Spark Open measured |
| LiveBench browser | Not measured | 36.0% | Only Muse Spark Open measured |
| LiveBench long agentic | Not measured | 18.9% | Only Muse Spark Open measured |
| Finance Agent (Vals AI) | 51.8% | 61.3% | Muse Spark Open |
| Tax Agent Bench (Vals AI) | 31.1% | 26.3% | Ember-1 |
| Legal Research Bench (Vals AI) | 44.7% | 38.5% | Ember-1 |
| Harvey's Legal Agent Benchmark (Vals AI) | Not measured | 25.0% | Only Muse Spark Open measured |
| Excel Modeling Benchmark (Vals AI) | 63.4% | 53.7% | Ember-1 |
| BioMysteryBench (Vals AI) | 67.0% | Not measured | Only Ember-1 measured |
| Terminal-Bench-Science (Vals AI) | 5.7% | Not measured | Only Ember-1 measured |
| Vals Science Index | 31.8% | Not measured | Only Ember-1 measured |
| APEX-Agents | Not measured | 33.4% | Only Muse Spark Open measured |
| Neither model has a result on the other 29 boards here | |||
ReasoningDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| LiveBench reasoning | Not measured | 73.7% | Only Muse Spark Open measured |
| MysteryMechanism (Vals AI) | 22.5% | Not measured | Only Ember-1 measured |
| GPQA Diamond (Artificial Analysis run) | Not measured | 89.0% | Only Muse Spark Open measured |
| Humanity's Last Exam (Artificial Analysis run) | Not measured | 43.9% | Only Muse Spark Open measured |
| Neither model has a result on the other 18 boards here | |||
MathNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| Neither model has a result on 13 boards here | |||
KnowledgeNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| Neither model has a result on 18 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index3 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| Design Arena: websites | Not measured | 1317 | Only Muse Spark Open measured |
| Design Arena: game dev | Not measured | 1294 | Only Muse Spark Open measured |
| LiveBench design | Not measured | 60.3% | Only Muse Spark Open measured |
| Neither model has a result on the other 6 boards here | |||
GeneralDirectional onlyPublic boards, shown as published, kept out of the index4 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | Leader |
|---|---|---|---|
| Vals Index | Not measured | 48.6% | Only Muse Spark Open measured |
| LiveBench data analysis | Not measured | 60.2% | Only Muse Spark Open measured |
| Public Benefits Bench (Vals AI) | 66.2% | Not measured | Only Ember-1 measured |
| Long context reasoning (AA-LCR) | Not measured | 78.3% | Only Muse Spark Open measured |
| Neither model has a result on the other 42 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | Ember-1 | Muse Spark Open | 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 Spark Open costs less
Repository review
50K input + 3K output tokens
Muse Spark Open costs less
Agent loop
200K input + 10K output tokens
Muse Spark Open 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 | Fireworks Ember-1 | Meta Muse Spark Open |
|---|---|---|
| Maker | Fireworks | Meta |
| Weights | No public weights listed | Open weights |
| Released | September 24, 2026 | October 15, 2026 |
| Context window | 1.05M | 1.05M |
| Input $/Mtok | $3.00 | $1.25 |
| Output $/Mtok | $15.00 | $4.25 |
| Coding index | 52.6 | 47.4 |
| Independent measurements behind it | 2 | 4 |
| Cost per solved task | Not computable | Not computable |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is Ember-1 better than Muse Spark Open for coding?
Pick Ember-1 for coding: it scores 52.6 on the Vibe Coding Index against 47.4 for Muse Spark Open, though on 2 benchmarks against 4. Pick Muse Spark Open for output tokens 3.5x cheaper ($4.25 vs $15.00 per million) and work measured by Finance Agent (Vals AI), where it leads.
Which is cheaper, Ember-1 or Muse Spark Open?
Ember-1 costs $3.00 per million input tokens and $15.00 per million output tokens; Muse Spark Open costs $1.25 and $4.25. Cost per solved task is known for neither of them.
Which has the bigger context window, Ember-1 or Muse Spark Open?
Neither: both take 1.05M tokens.
How many benchmarks is the Ember-1 vs Muse Spark Open comparison based on?
7 benchmarks have a result for both models as of October 21, 2026. Ember-1's coding index rests on 2 independent benchmarks and Muse Spark Open's on 4. Where only one model has a result, the page says so instead of naming a winner.