Grok 4.20 vs Muse Spark Open 816B A42B
Full profiles: Grok 4.20 (SpaceXAI) and Muse Spark Open 816B A42B (Meta): every benchmark, price and rank.
Updated · Data from 10 sources
As of 22 Nov 2026, Grok 4.20 and Muse Spark Open 816B A42B are neck and neck on quality (58.2 vs 57.8 on the BenchLeader Index, inside the ±7.3 margin), and Grok 4.20 is 1.3× cheaper at $1.56 vs $2.00 per 1M tokens blended. Muse Spark Open 816B A42B leads in every category we measure both in.
Key facts (as of 22 Nov 2026)
- Grok 4.20 scores 58.2 on the BenchLeader Index and Muse Spark Open 816B A42B scores 57.8 (BenchLeader, 22 Nov 2026).
- Grok 4.20 costs $1.56 per 1M tokens blended (list price, 22 Nov 2026).
- Muse Spark Open 816B A42B costs $2.00 per 1M tokens blended (list price, 22 Nov 2026).
- Of the 5 benchmarks both models have, Muse Spark Open 816B A42B leads 4 (BenchLeader, 22 Nov 2026).
Which to pick for…
The pick's numbers come first. Category scores on the BenchLeader scale (50 is the average model, 15 points one standard deviation). High confidence means a lead of half a standard deviation or more; under one point is called either.
| Use case | Pick | Why | Confidence |
|---|---|---|---|
| Coding | Muse Spark Open 816B A42B | 65 vs 53 category score | High |
| Maths | Muse Spark Open 816B A42B | 58 vs 54 category score; FrontierMath Tiers 1–3 69.8% vs 44.9% | Medium |
| Reasoning | Muse Spark Open 816B A42B | 63 vs 60 category score; GPQA Diamond 88.6% vs 89.3% | Medium |
| Knowledge | Muse Spark Open 816B A42B | 61 vs 53 category score; SimpleQA Verified 56.6% vs 30.2% | High |
| Chat quality (human preference) | – | Only Grok 4.20 has an Arena rating (66 category score) | Not comparable |
| Lowest cost | Grok 4.20 | $1.56 vs $2.00 per 1M blended; agentic coding session $0.085 vs $0.092 | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Grok 4.20 (thinking) and Muse Spark Open 816B A42B (xhigh) are neck and neck on quality (58.2 vs 57.8, within the ±7.3 margin).
- Muse Spark Open 816B A42B (xhigh) is stronger in coding, knowledge, maths, reasoning.
- Grok 4.20 (thinking) is 1.3× cheaper ($1.56 vs $2.00 per 1M blended).
At a glance
| Spec | Grok 4.20 | Muse Spark Open 816B A42B |
|---|---|---|
| Developer | SpaceXAI | Meta |
| Released | 17 Feb 2026 | 15 Oct 2026 |
| Context window (tokens) | 1,000,000 | 1,048,576 |
| Compared at | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | $1.25 / $2.50 / $0.200 | $1.25 / $4.25 / $0.150 |
Category by category
Category scores on the BenchLeader scale: 50 is the average evaluated model, 15 points is one standard deviation. A gap under one point is called level.
| Category | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) | Leader |
|---|---|---|---|
| Reasoning | 59.9 | 63.1 | Muse Spark Open 816B A42B |
| Coding | 53.2 | 65.1 | Muse Spark Open 816B A42B |
| Agents & tools | 50.5 | – | not comparable |
| Maths | 53.6 | 58.2 | Muse Spark Open 816B A42B |
| Knowledge | 52.9 | 60.5 | Muse Spark Open 816B A42B |
| Instruction following | 78.8 | – | not comparable |
| Human preference | 65.9 | – | not comparable |
| Multimodal | 58.2 | – | not comparable |
| Long context | 58.9 | – | not comparable |
On one chart
Each domain on the index's own scale, where 50 is an average model. Click a domain to see the benchmarks behind it. Pick up to 5 models from the list.
Not drawn, measured for fewer than two of these models: Agents & tools, Instruction following, Long context, Multimodal, Human preference.
Where sources disagree
Across the 5 benchmarks both have, Grok 4.20 leads 1, Muse Spark Open 816B A42B leads 4.
| Overall index | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) | Leader |
|---|---|---|---|
| BenchLeader Index | 58.2 | 57.8 | neck and neck |
Each publisher runs its own tasks, harnesses and effort settings, so their overall verdicts can differ; BenchLeader reconciles them on one scale. † measured at another effort setting of the same model (the publisher did not test the one compared here).
Price, speed and latency
List prices blended at three input tokens per output token. Per-task prices use the token counts on the methodology page. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) |
|---|---|---|
| Blended price per 1M tokens | $1.56 | $2.00 |
| Input / output per 1M | $1.25 / $2.50 | $1.25 / $4.25 |
| Cached input per 1M | $0.200 | $0.150 |
| Price per task: chat reply | $0.0013 | $0.0018 |
| Price per task: summarize a 30-page report | $0.017 | $0.018 |
| Price per task: code edit | $0.011 | $0.014 |
| Price per task: agentic coding session | $0.085 | $0.092 |
| Price per task: structured extraction | $0.0030 | $0.0034 |
| Agentic coding session, input mostly cached | $0.038 | $0.043 |
| Output speed | 132 tok/s | – |
| Time to first answer | 9.2 s | – |
| Time to a complete answer, last measured 13 Nov 2026 | 43.8 s | – |
Each model is compared at its best-ranked reasoning-effort setting. Grok 4.20 (thinking) scores 58.2 and takes 9 s to its first answer token. Muse Spark Open 816B A42B (xhigh) scores 57.8. Higher effort means more thinking before the answer: better scores, longer waits and more tokens billed. How reasoning effort changes scores and cost.
Every benchmark
78 benchmarks, grouped by category; each name links to the board and its leaders. † measured at another effort setting of the same model, shown for reference and not counted as a win.
| Metric | Publisher | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 58.2 | 57.8 |
| Agents & tools score | BenchLeader | 50.5 | – |
| Coding score | BenchLeader | 53.2 | 65.1 |
| Human preference score | BenchLeader | 65.9 | – |
| Instruction following score | BenchLeader | 78.8 | – |
| Knowledge score | BenchLeader | 52.9 | 60.5 |
| Long context score | BenchLeader | 58.9 | – |
| Maths score | BenchLeader | 53.6 | 58.2 |
| Multimodal score | BenchLeader | 58.2 | – |
| Reasoning score | BenchLeader | 59.9 | 63.1 |
| Blended price $/M | BenchLeader | $1.56 | $2.00 |
| Output speed | BenchLeader | 132 tok/s | – |
| Time to first answer | BenchLeader | 9.2 s | – |
| Context window | BenchLeader | 1,000,000 tokens | 1,048,576 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 89.3% | 88.6% |
| Arena Hard Prompts | Arena (LMArena) | 1489 | – |
| GPQA Diamond (AA) | Artificial Analysis | 91.1% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 34.5% | – |
| GPQA Diamond (Vals) | Vals AI | 88.6% | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 75.0% | – |
| ARC-AGI-1 | ARC Prize | 89.5% | – |
| ARC-AGI-2 | ARC Prize | 65.1% | – |
| ARC-AGI-3 | ARC Prize | 0.1% | – |
| CritPt | Artificial Analysis | 6.6% | – |
| Arena Expert | Arena (LMArena) | 1477 | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | 29.0% | – |
| CL-bench | Epoch AI Benchmarking Hub | 22.2%† | – |
| CL-bench Life | Epoch AI Benchmarking Hub | 11.9%† | – |
| LMCA | Epoch AI Benchmarking Hub | 38.7%† | – |
| DTBench | Epoch AI Benchmarking Hub | 90.1%† | – |
| ForecastBench | ForecastBench | 64.0% | – |
| Coding | |||
| SciCode | SciCode | – | 58.5% |
| WeirdML | WeirdML | 52.3%† | – |
| Arena Coding | Arena (LMArena) | 1511 | – |
| Arena WebDev | Arena (LMArena) | 1376 | – |
| LiveCodeBench | Vals AI | 84.3% | – |
| SWE-bench (Vals) | Vals AI | 72.2% | – |
| Code Migration | Vals AI | 0.3% | – |
| Vibe Code Bench v1.1 | Vals AI | 4.1% | – |
Show all 78 benchmarks
| Metric | Publisher | Grok 4.20 (thinking) | Muse Spark Open 816B A42B (xhigh) |
|---|---|---|---|
| Coding | |||
| ALE-Bench | Epoch AI Benchmarking Hub | 1150.3† | 1590.1 |
| Design Arena Code | Design Arena | 1202 | – |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 57.3%† | – |
| APEX-Agents | Mercor | – | 33.4%† |
| Terminal-Bench Hard | Artificial Analysis | 37.9% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 93.0% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 44.2% | – |
| HiL-Bench | Scale AI SEAL | 20.0%† | – |
| APEX-Agents (AA) | Artificial Analysis | 14.2% | – |
| Finance Agent v2 | Vals AI | 28.5% | – |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 0.0% | – |
| Legal Research Bench | Vals AI | 13.9% | – |
| Tax Agent Bench | Vals AI | 7.0% | – |
| Terminal-Bench 2.0 (Vals) | Vals AI | 40.5% | – |
| Vending-Bench 2 | Andon Labs | 4662.9† | – |
| Design Arena Agentic | Design Arena | 1083 | – |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 44.9% | 69.8% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 17.1% | 31.7% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 92.2% | 98.1% |
| ProofBench | Vals AI | 14.0% | – |
| AIME (Vals) | Vals AI | 96.5% | – |
| Arena Maths | Arena (LMArena) | 1464 | – |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 30.2% | 56.6% |
| MMLU-Pro | Vals AI | 86.3% | – |
| LegalBench | Vals AI | 77.7% | – |
| CorpFin | Vals AI | 63.7% | – |
| TaxEval | Vals AI | 74.1% | – |
| MedQA | Vals AI | 94.5% | – |
| CaseLaw v2 | Vals AI | 54.5% | – |
| Excel Modeling Benchmark | Vals AI | 11.7% | – |
| MedCode | Vals AI | 32.2% | – |
| MedScribe | Vals AI | 63.4% | – |
| Arena Text Factuality | Arena (LMArena) | 1456 | – |
| Global-MMLU-Lite (AA) | Artificial Analysis | 89.5% | – |
| Instruction following | |||
| IFBench | Artificial Analysis | 81.2% | – |
| Arena Instruction Following | Arena (LMArena) | 1448 | – |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1472 | – |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1574† | – |
| EQ-Bench Longform Writing | EQ-Bench | 68.5† | – |
| Arena Creative Writing | Arena (LMArena) | 1445 | – |
| Arena Multi-turn | Arena (LMArena) | 1479 | – |
| Arena Longer Queries | Arena (LMArena) | 1466 | – |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1263 | – |
| MMMU-Pro | Artificial Analysis | 74.6% | – |
| MMMU-Pro (Vals) | Vals AI | 83.5% | – |
| MortgageTax | Vals AI | 45.4% | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 0.0%† | – |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 69.0% | – |
| Arena Document | Arena (LMArena) | 1440 | – |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 152.3† | – |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 25.7 | – |
| AA Progress Index v1 | Artificial Analysis | 281.9 | – |
| Vals Multimodal Index | Vals AI | 39.1% | – |
Data as of 2026-11-22. Best configuration of each model; every score links to its source on the model pages.
What changed
- Grok 4.20 (thinking): 1 score revision.
- Muse Spark Open 816B A42B (xhigh): BenchLeader Index 58.6 → 58.1.
- Grok 4.20 (thinking): BenchLeader Index 58.9 → 58.4.
- Grok 4.20 (thinking): 1 new result (AA Progress Index v1).
- Grok 4.20 (thinking): 1 new result (Global-MMLU-Lite (AA)).
- Grok 4.20 (thinking): 1 new result (Harvey's Legal Agent Benchmark v1).
Related comparisons
- Grok 4.20 vs Gemini 4 Carbon
- Grok 4.20 vs GPT-6 Astra
- Muse Spark Open 816B A42B vs Claude Fable 5.2
- Muse Spark Open 816B A42B vs Gemini 4 Carbon
- Muse Spark Open 816B A42B vs GPT-6 Astra
- Muse Spark Open 816B A42B vs Claude Opus 5.5
- Grok 4.20 vs GPT-6.1 Sol
- Grok 4.20 vs Gemini 4 Argon
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Grok 4.20 vs Muse Spark Open 816B A42B: questions
- Is Grok 4.20 better than Muse Spark Open 816B A42B?
- Grok 4.20 (thinking) and Muse Spark Open 816B A42B (xhigh) are neck and neck on quality (58.2 vs 57.8, within the ±7.3 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 22 Nov 2026, so check the category scores for your use.
- Grok 4.20 vs Muse Spark Open 816B A42B: which is cheaper, and will that change?
- Grok 4.20 is cheaper today: $1.56 against $2.00 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Grok 4.20 or Muse Spark Open 816B A42B better for coding?
- Muse Spark Open 816B A42B scores higher in coding (65 vs 53 on the category index, where 50 is average).
- Which is better for long documents?
- Both take about 1M tokens: Grok 4.20 1,000,000, Muse Spark Open 816B A42B 1,048,576.
- Do independent benchmarks agree on Grok 4.20 vs Muse Spark Open 816B A42B?
- Across the 5 benchmarks both have, Grok 4.20 leads 1, Muse Spark Open 816B A42B leads 4.
More questions (1)
- When was Muse Spark Open 816B A42B released?
- Muse Spark Open 816B A42B was released on 15 Oct 2026.