Compare / Muse Spark 1.3 vs MiniMax M3
Muse Spark 1.3 vs MiniMax M3
Which is cheaper, which is more capable, and which does better at each kind of work, from 24 results on 3 sources that measured both models.
Muse Spark 1.3 or MiniMax M3? The short answer
- MiniMax M3 is cheaper: $0.52 against $2.00 per million tokens, 3.8 times less.
- Muse Spark 1.3 scores higher on the Artificial Analysis Intelligence Index: 48.1 against 29.2.
- Muse Spark 1.3 writes faster: 137 tokens a second against 81.
- On our business benchmarks, Muse Spark 1.3 ranks higher for agents and tool use, professional work, reasoning and knowledge, writing and sticking to the facts.
At a glance
| Muse Spark 1.3 | MiniMax M3 | |
|---|---|---|
| Price per million tokens | $1.25 in · $4.25 out | $0.30 in · $1.20 out |
| Blended (3 in : 1 out) | $2.00 | $0.52 |
| Cheapest host, blended | – | $0.41 (CoreWeave, FP4) |
| Intelligence Index | 48.1 | 29.2 |
| Spring Prompt overall | 66 | – |
| Output speed, tokens a second | 137 | 81 |
| Context window | 1,048,576 tokens | 1,048,576 tokens |
| Weights | Closed | Open |
| Developer based in | the US | China |
Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.
Which is better for what
Each model's rank among models available today, for each kind of work, from our own benchmarks and licensed sources. The better rank is in green.
| Use case | Muse Spark 1.3 | MiniMax M3 |
|---|---|---|
| Product listingsTurning a sparse product feed and photos into listings that can go live | 9 of 19 | not measured |
| Decks from an analysisTurning a finished analysis into a deck you could present as it is | 13 of 18 | not measured |
| Marketing planningPlanning a year of ad spend without overspending | 12 of 18 | not measured |
| Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls | 9 of 161 | 71 of 161 |
| Professional workReal tasks from banking, consulting and law, business documents and freelance projects | 8 of 51 | 39 of 51 |
| Reasoning and knowledgeHard questions across science, maths and general knowledge | 8 of 195 | 36 of 195 |
| WritingWhat people prefer in blind comparisons, and judged writing quality | 7 of 187 | 61 of 187 |
| Sticking to the factsSummarising without inventing things, and factual answers | 10 of 153 | 94 of 153 |
| Speed under pressureGood decisions against a real clock (fast chess) | 18 of 21 | not measured |
From our research
Every shared result
| Benchmark | Metric | Muse Spark 1.3 | MiniMax M3 |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 57.8% | 37.7% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.08 | -0.14 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.04 | -0.13 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.02 | -0.07 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | -0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,468 | 1,433 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,479 | 1,434 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,506 | 1,450 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,450 | 1,406 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,527 | 1,471 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,486 | 1,435 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,494 | 1,440 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,466 | 1,422 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 76.5 | 58.6 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 48.1 | 29.2 |
| Artificial Analysisreported | GPQA Diamond% of questions | 94.1% | 92.9% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 48.7% | 39.0% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 83.0% | 83.0% |
| Artificial Analysisreported | Output speedtokens per second | 137 | 80.6 |
| Artificial Analysisreported | SciCode% of problems | 59.7% | 47.1% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 85.4% | 65.2% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 33.3% | 2.0% |
| Artificial Analysisreported | Time to first answer tokenseconds | 41.3 | 25.9 |
| Artificial Analysisreported | Τ-bench banking% of tasks | 50.5% | 15.3% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.
Quick answers
Is Muse Spark 1.3 better than MiniMax M3?
It depends on the task. Among models available today, on our benchmarks and licensed sources, Muse Spark 1.3 ranks higher for agents and tool use, professional work, reasoning and knowledge, writing and sticking to the facts.
Which is cheaper, Muse Spark 1.3 or MiniMax M3?
MiniMax M3 costs $0.52 per million tokens (three input to one output), against $2.00 for Muse Spark 1.3.
Which is faster, Muse Spark 1.3 or MiniMax M3?
Muse Spark 1.3 writes about 137 tokens a second on its usual API, against 81 for MiniMax M3 (Artificial Analysis).
Run Muse Spark 1.3 and MiniMax M3 on your own prompt
Benchmarks aren't your data. Try both side by side in the playground, with the cost of every answer.