MetaMuse Spark 1.3
Best configuration ranks #41 of 610 on the BenchLeader Index at 63.9 ±5.4 (xhigh reasoning effort). Last measured 8 Sept 2026. Released 2 Sept 2026.
- Blended price
- $2.00/M
- $1.25 in · $4.25 out
- Output speed
- 193 tok/s
- First answer
- 40 s
- first token 3.68 s
- Context
- 1M
- Overall index64
- Coding71
- Agents & tools60
- Knowledge75
- Multimodal67
- Long context68
- Composite79
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Default” means the publisher did not say which setting was used.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| xhighbest | 63.9 | #41 | 193 tok/s | 40 s | $0.0018 | Agents & tools 60 · Coding 71 · Composite 79 · Knowledge 75 · Long context 68 · Multimodal 67 |
| max | – | – | 226 tok/s | 36 s | $0.0018 | Agents & tools 67 · Coding 74 |
| default | – | – | 226 tok/s | 36 s | $0.0018 | Composite 90 · Knowledge 76 · Long context 68 |
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Reasoningnot in index | 89.7%#9 | – | – | LiveBench | |
| GPQA Diamond (AA)not in index | 94.1%#7 | – | 93.5%#14 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 47.5%#21 | – | 48.7%#17 | Artificial Analysis |
Coding
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LMArena WebDev | 1625#10 | 1650#7 | – | LMArena | |
| LiveBench Codingnot in index | 81.1%#11 | – | – | LiveBench | |
| SciCode (AA)not in index | 59.7%#5 | – | 58.8%#8 | Artificial Analysis |
Agents & tools
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Agentic Codingnot in index | 64.1%#5 | – | – | LiveBench | |
| Terminal-Bench 2.1 (Vals) | 72.3%#17 | 79.0%#8 | – | Vals AI |
Maths
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Mathematicsnot in index | 96.0%#5 | – | – | LiveBench |
Knowledge
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Data Analysisnot in index | 79.6%#6 | – | – | LiveBench | |
| AA-Omniscience | 23.1#33 | – | 25#31 | Artificial Analysis |
Instruction following
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench Languagenot in index | 82.8%#17 | – | – | LiveBench |
Multimodal
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| MMMU-Pro | 82.0%#22 | – | – | Artificial Analysis |
Long context
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| AA-LCR | 83.0%#12 | – | 83.0%#12 | Artificial Analysis |
Composite
| Benchmark | xhigh | max | default | Source | Trend |
|---|---|---|---|---|---|
| LiveBench | 81.6%#4 | – | – | LiveBench | |
| AA Intelligence Index | 45.2#18 | – | 48.2#13 | Artificial Analysis | |
| Vals Indexnot in index | 60.3#9 | 64.5#5 | – | Vals AI |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Meta | 76 tok/s | 3.68 s | $1.25 | $4.25 | 1.0M | – |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.150 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0018 | $0.0014 | 41.5 s |
| Summarise a 30-page report | 12,000 / 600 | $0.018 | $0.0077 | 43.1 s |
| Code edit | 6,000 / 1,500 | $0.014 | $0.0089 | 47.7 s |
| Agentic coding session | 60,000 / 4,000 | $0.092 | $0.043 | 1.0 min |
| Structured extraction | 2,000 / 200 | $0.0034 | $0.0017 | 41.0 s |
See also
Data as of 9 Sept 2026. Compare these configurations.