Quality
Highest overall scores
- 1
Meta: Muse Spark 1.150.6 - 2Meta: Llama 4 Maverick14.3
- 3Meta: Llama 3.3 70B Instruct9.4
Meta model benchmarks.
Compare Meta models across answer quality, coding, complex tasks, speed, price, and context size.
Response speed
Fastest output in this sample
- 1Meta: Muse Spark 1.1143 t/s
- 2Meta: Llama 3.3 70B Instruct104 t/s
- 3Meta: Llama 4 Maverick55 t/s
Input price
Lowest price per 1M tokens
- 1Meta: Llama 3.3 70B Instruct$0.10/1M
- 2Meta: Llama 4 Maverick$0.20/1M
- 3Meta: Muse Spark 1.1$1.25/1M
Highlights
Quality
Overall capability · Higher is better
Speed
Output tokens per second · Higher is better
Input price
USD per 1M tokens · Lower is better
Quality benchmarks
Compare Meta model capability.
Performance and price
Compare practical tradeoffs.
All Meta models.
3 models in the current catalogue.
| Model | Links | |||||
|---|---|---|---|---|---|---|
| Meta: Muse Spark 1.1 | 50.6 | 143 t/s | 1M | $1.25 | $4.25 | |
| Meta: Llama 4 Maverick | 14.3 | 55 t/s | 1M | $0.20 | $0.80 | |
| Meta: Llama 3.3 70B Instruct | 9.4 | 104 t/s | 131.1K | $0.10 | $0.32 |
Meta technical benchmarks.
Compare named evaluation records available for Meta models. Missing results remain blank rather than being estimated.
| Benchmark | Area | Muse Spark 1.1 | Llama 4 Maverick | Llama 3.3 70B Instruct | Source |
|---|---|---|---|---|---|
| Intelligence Index | overallQuality | 50.6 | 14.3 | 9.4 | OpenRouter / artificial-analysis |
| Coding Index | coding | 71.3 | 16.3 | 11.9 | OpenRouter / artificial-analysis |
| Agentic Index | agentic | 37.5 | 1.3 | 0.3 | OpenRouter / artificial-analysis |
| GPQA | GPQA Diamond | 89.8 | 67.1 | 49.8 | OpenRouter model benchmarks |
| Humanity's Last Exam | HLE | 45.1 | 4.8 | 4 | OpenRouter model benchmarks |
| AA-LCR | AA-LCR | 63.3 | 46 | 15 | OpenRouter model benchmarks |
| GDPval-AA | GDPval-AA | 43.8 | 0 | 0 | OpenRouter model benchmarks |
| CritPt | CritPt | 15.1 | 0 | 0 | OpenRouter model benchmarks |
| SciCode | SciCode | 58.2 | 33.1 | 26 | OpenRouter model benchmarks |
| AA-Omniscience Accuracy | AA-Omniscience Accuracy | 40.6 | 24.3 | 17.9 | OpenRouter model benchmarks |
| AA-Omniscience Non-Hallucination Rate | AA-Omniscience Non-Hallucination Rate | 61.9 | 12.7 | 14.9 | OpenRouter model benchmarks |
| IFBench | IFBench | — | 43 | 47.1 | OpenRouter model benchmarks |
| τ²-Bench Telecom | τ²-Bench Telecom | — | 17.8 | 26.6 | OpenRouter model benchmarks |
| Terminal-Bench Hard | Terminal-Bench Hard | — | 6.8 | 3 | OpenRouter model benchmarks |
| SWE-bench Pro | Coding | — | 5.24 | — | OpenEvals/leaderboard-data |