GPT-6 Astra vs Muse Spark 1.3
Wins 2 of 8 areas
Reasoning · Math
Wins 2 of 8 areas
Long documents · Following instructions
The two are evenly matched.GPT-6 Astra is better at reasoning and math; Muse Spark 1.3 at long documents and following instructions.
Scores updated · 32 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking50GPT-6 Astra5 of 5 tests
- MathCompetition and research-level math20GPT-6 Astra2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01Muse Spark 1.31 of 1 test
- Following instructionsDoing exactly what it is asked01Muse Spark 1.31 of 1 test
- CodingWriting and fixing software22Even2 each · 1 tie
- AgentsCarrying out multi-step tasks on its own22Even2 each
- FactsGetting facts right instead of making them up11Even1 each
- Images and chartsUnderstanding pictures, charts and video11Even1 each
Long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
Muse Spark 1.3 costs 91% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-6 Astra pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+51.3points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+25.8points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+19.7points ahead
Where Muse Spark 1.3 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+18.4points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+9.1points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+6.8points ahead
Every test, side by side
All 32 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- LMArena · WebDevGPT-6 Astra by 115 rating points17741659+115 rating
- LiveBench · Agentic CodingMuse Spark 1.3 by 6.857.364.1+6.8
- Terminal-Bench 2.1GPT-6 Astra by 4.188.484.3+4.1
- SciCodeMuse Spark 1.3 by 2.356.558.8+2.3
- LiveBench · Codingtie80.481.1tie
AgentsEven
- Terminal-Bench 4.0GPT-6 Astra by 25.859.133.3+25.8
- AA IT-Bench SREGPT-6 Astra by 15.448.633.2+15.4
- τ-Bench V3 · BankingMuse Spark 1.3 by 9.141.450.5+9.1
- GDPValMuse Spark 1.3 by 5.453.659+5.4
ReasoningGPT-6 Astra
- CritPtGPT-6 Astra by 6.831.724.9+6.8
- Humanity's Last ExamGPT-6 Astra by 654.748.7+6
- LiveBench · ReasoningGPT-6 Astra by 392.789.7+3
- GPQA DiamondGPT-6 Astra by 2.696.193.5+2.6
- SimpleBenchGPT-6 Astra by 1.883.681.8+1.8
FactsEven
- AA-Omniscience · AccuracyGPT-6 Astra by 1962.643.6+19
- AA-Omniscience · Non-hallucinationMuse Spark 1.3 by 18.448.767.1+18.4
Images and chartsEven
- MMMU-ProGPT-6 Astra by 4.986.982+4.9
- LMArena · VisionMuse Spark 1.3 by 28 rating points12801308+28 rating
MathGPT-6 Astra
- FrontierMath Tier 4GPT-6 Astra by 51.397.646.3+51.3
- FrontierMath Tiers 1-3 (v2)GPT-6 Astra by 19.793.774+19.7
- LiveBench · Mathematicstie96.896tie
Following instructionsMuse Spark 1.3
- LiveBench · Instruction FollowingMuse Spark 1.3 by 2.475.678+2.4
Other results9 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- AA-OmniscienceGPT-6 Astra by 18.443.425+18.4
- livebench_languageGPT-6 Astra by 6.689.482.8+6.6
- OSWorld 2.0GPT-6 Astra by 5.772.666.9+5.7
- AA IntelligenceGPT-6 Astra by 4.652.748.1+4.6
- AA Agentic IndexMuse Spark 1.3 by 4.251.555.7+4.2
- livebench_data_analysisGPT-6 Astra by 3.48379.6+3.4
- HiL-Bench (Tools-allowed)Muse Spark 1.3 by 1.665.366.9+1.6
- DeepSWE 1.1Muse Spark 1.3 by 1.374.175.4+1.3
- Artificial Analysis Coding IndexGPT-6 Astra by 1.176.975.8+1.1
Questions people ask
Which is better, GPT-6 Astra or Muse Spark 1.3?
GPT-6 Astra and Muse Spark 1.3 each win two of the eight areas we test. GPT-6 Astra wins reasoning and math; Muse Spark 1.3 wins long documents and following instructions. Muse Spark 1.3 costs 91% less. They are level on coding, agents, facts and images and charts.
Which is better for coding?
Neither. They win 2 coding tests each of the 5 both models report, and 1 is a tie.
Which is cheaper?
GPT-6 Astra costs $10.00 per million input tokens and $50.00 per million output tokens; Muse Spark 1.3 costs $1.25 and $4.25. That makes Muse Spark 1.3 about 91% cheaper for the same work.
How do you compare the two?
We use the 32 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.