vs
0:10
overall capability wins (0 ties across 10 scored). Muse Spark 1.2 leads overall.
- Reasoning0:4
- Coding0:2
- Arena0:1
- Tool use0:1
- Composite indices0:2
- Pricing3:0
- Speed1:0
- Specs0:2
Verdict
Muse Spark 1.2 leads coding (widest gap: +72.2 on TermBench); Muse Spark 1.2 ranks higher on LMArena (+173 Elo); Muse Spark 1.2 offers larger context (1.0M vs 1M).
- Muse Spark 1.2 leads coding (widest gap: +72.2 on TermBench)
- Muse Spark 1.2 ranks higher on LMArena (+173 Elo)
- Muse Spark 1.2 offers larger context (1.0M vs 1M)
- Muse Spark 1.2 allows ~4.0x more max output (131K vs 33K)
- Llama 4 Maverick ships open weights (Llama 4 Community)
- Llama 4 Maverick is ~4.8x cheaper on a blended token basis than Muse Spark 1.2
Cost for 1M in + 250K out
Illustrative chat workload at primary-provider list prices.
- Llama 4 Maverick
- $0.48
- Muse Spark 1.2
- $2.31
Llama 4 Maverick is about 4.8x cheaper on this workload.
Price per 1M tokens
Full comparison
Organization
Meta
Meta
Family
Llama / Muse
Llama / Muse
License
Llama 4 Community
Proprietary
Open weights
Yes
No
Release
Apr 5, 2025
Aug 5, 2026
Knowledge cutoff
-
-
API / provider
Together / Fireworks (ref.)
Meta
Modalities
text, image → text
text, image, video, audio, pdf → text
Specs
Context window
1M
1.0M
Max output
33K
131K
Parameters
400B (17B act.)
—
Pricing
Input $/1M
$0.27
$1.25
Output $/1M
$0.85
$4.25
Blended $/1M (3∶1)
$0.42
$2
Speed
tok/s
95
95
TTFT (s)
0.35
0.38
Reasoning
MMLU-Pro
80.5
—
GPQA Diamond
67.1
90.4
Humanity's Last Exam
4.9
45.5
AIME 2025
19.3
—
MATH-500
88.9
—
Humanity's Last Exam (with tools)
—
—
AA-Omniscience Accuracy
24.9
45.4
AA-LCR v1.1
50.0
79.0
CritPt
—
17.7
MMMU-Pro
62.1
—
IFBench
43.0
—
Chartography
—
—
Chartography (With Tools)
—
—
Coding
SWE-bench Verified
21.0
—
SWE-bench Pro
5.2
—
SWE-bench Multilingual
—
—
LiveCodeBench
43.4
—
Terminal-Bench 2.1
7.9
80.1
Aider Polyglot
15.6
—
Terminal-Bench 3
—
—
BigCodeBench
49.7
—
SciCode
31.7
57.4
CursorBench
—
—
SWE-Rebench
11.0
—
NL2Repo-Bench
—
—
DeepSWE
—
55.0
WebDev Arena
—
1,535
Terminal-Bench 4.0
—
—
CursorBench 4.0
—
—
FrontierCode v1.1 (Main)
—
—
Arena
LMArena Elo
1,327
1,500
| Metric | Llama 4 Maverick | Muse Spark 1.2 | Delta |
|---|---|---|---|
| Identity | |||
| Organization | Meta | Meta | — |
| Family | Llama / Muse | Llama / Muse | — |
| License | Llama 4 Community | Proprietary | — |
| Open weights | Yes | No | — |
| Release | Apr 5, 2025 | Aug 5, 2026 | — |
| Knowledge cutoff | - | - | — |
| API / provider | Together / Fireworks (ref.) | Meta | — |
| Modalities | text, image → text | text, image, video, audio, pdf → text | — |
| Specs | |||
| Context window | 1M | 1.0M | B +49K |
| Max output | 33K | 131K | B +98K |
| Parameters | 400B (17B act.) | — | — |
| Pricing | |||
| Input $/1M | $0.27 | $1.25 | A +$0.98 |
| Output $/1M | $0.85 | $4.25 | A +$3.40 |
| Blended $/1M (3∶1) | $0.42 | $2 | A +$1.58 |
| Speed | |||
| tok/s | 95 | 95 | tie |
| TTFT (s) | 0.35 | 0.38 | A +0.03 |
| Reasoning | |||
| MMLU-Pro | 80.5 | — | — |
| GPQA Diamond | 67.1 | 90.4 | B +23.3 pts |
| Humanity's Last Exam | 4.9 | 45.5 | B +40.6 pts |
| AIME 2025 | 19.3 | — | — |
| MATH-500 | 88.9 | — | — |
| Humanity's Last Exam (with tools) | — | — | — |
| AA-Omniscience Accuracy | 24.9 | 45.4 | B +20.5 pts |
| AA-LCR v1.1 | 50.0 | 79.0 | B +29.0 pts |
| CritPt | — | 17.7 | — |
| MMMU-Pro | 62.1 | — | — |
| IFBench | 43.0 | — | — |
| Chartography | — | — | — |
| Chartography (With Tools) | — | — | — |
| Coding | |||
| SWE-bench Verified | 21.0 | — | — |
| SWE-bench Pro | 5.2 | — | — |
| SWE-bench Multilingual | — | — | — |
| LiveCodeBench | 43.4 | — | — |
| Terminal-Bench 2.1 | 7.9 | 80.1 | B +72.2 pts |
| Aider Polyglot | 15.6 | — | — |
| Terminal-Bench 3 | — | — | — |
| BigCodeBench | 49.7 | — | — |
| SciCode | 31.7 | 57.4 | B +25.7 pts |
| CursorBench | — | — | — |
| SWE-Rebench | 11.0 | — | — |
| NL2Repo-Bench | — | — | — |
| DeepSWE | — | 55.0 | — |
| WebDev Arena | — | 1,535 | — |
| Terminal-Bench 4.0 | — | — | — |
| CursorBench 4.0 | — | — | — |
| FrontierCode v1.1 (Main) | — | — | — |
| Arena | |||
| LMArena Elo | 1,327 | 1,500 | B +173 Elo |
Llama 4 Maverick
- Context
- 1M / 33K out
- Parameters
- 400B (17B act.)
- Price
- $0.27 / $0.85
- Speed
- 95 tok/s · 0.35s TTFT
- Modalities
- text, image → text
- License
- Llama 4 Community
Meta's open multimodal MoE flagship - 400B total / 17B active with 1M context.
Muse Spark 1.2
- Context
- 1.0M / 131K out
- Parameters
- —
- Price
- $1.25 / $4.25
- Speed
- 95 tok/s · 0.38s TTFT
- Modalities
- text, image, video, audio, pdf → text
- License
- Proprietary
Co-trained with Muse Code for higher first-attempt accuracy and more reliable tool calling — 1M context coding-focused upgrade; Meta Model API at $1.25/$4.25 (contributor tier $0.10/$0.20).
Benchmark charts
Winner bars are emphasized. Per-benchmark deltas sit above each chart.