SpaceXAIGrok 4.1 Fast
Grok 4.1 Fast is a SpaceXAI proprietary reasoning model, released 19 Nov 2025. Its best configuration (thinking reasoning effort) ranks #234 of 777 on the BenchLeader Index at 54.8 ±3.7, in the upper half. The ± is the point: 267 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in agents & tools (71) and lowest in coding (46). At $0.275 per million tokens blended it is cheaper than most ranked models. It has been measured at 3 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 20 Oct 2026.
- Blended price
- $0.275/M
- $0.200 in · $0.500 out
- Output speed
- –
- First answer
- –
- Context
- 128k
- Full answer
- –
- Cost per run
- –
- Released
- 19 Nov 2025
- Overall index55
- Reasoning53
- Coding46
- Agents & tools71
- Maths50
- Knowledge55
- Instruction following54
- Human preference61
- Multimodal49
- Long context62
- Composite52
Versions
SpaceXAI has shipped 2 models under this name. Each is ranked on its own results; a newer version often has fewer results so far, which holds its index nearer the average until more arrive.
| Model | Released | Index | Rank |
|---|---|---|---|
| Grok 4.1 Fastthinkingthis page | 19 Nov 2025 | 54.8 | #234 |
| Grok 4 Fastthinking | 19 Sept 2025 | 51.5 | #339 |
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| no reasoning | 42.6 | #581 | – | – | $0.0002 | Agents & tools 61 · Coding 28 · Composite 42 · Instruction following 40 · Knowledge 41 · Long context 40 · Maths 40 · Multimodal 31 · Reasoning 48 |
| high | – | – | – | – | – | |
| thinkingbest | 54.8 | #234 | – | – | $0.0002 | Agents & tools 71 · Coding 46 · Composite 52 · Human preference 61 · Instruction following 54 · Knowledge 55 · Long context 62 · Maths 50 · Multimodal 49 · Reasoning 53 |
How its index has moved
8 Sept 2026 to 18 Oct 2026- thinking
- no reasoning
The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| SimpleBench | 56.0%#37 | 56.0%#37 | SimpleBench | |
| Arena Hard Prompts | – | 1441#122 | Arena (LMArena) | |
| GPQA Diamond (AA)not in index | 63.7%#344 | 85.3%#125 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 5.1%#438 | 19.3%#215 | Artificial Analysis | |
| GPQA Diamond (Vals)not in index | 65.2%#98 | 84.3%#51 | Vals AI | |
| CritPt | 0.0%#307 | 2.9%#177 | Artificial Analysis | |
| DTBenchnot in index | – | 87.7%#97 | Epoch AI Benchmarking Hub | |
| ForecastBenchnot in index | – | 64.8%#19 | ForecastBench |
Coding
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Arena Coding | – | 1461#127 | Arena (LMArena) | |
| Arena WebDev | – | 1242#126 | Arena (LMArena) | |
| LiveCodeBench | 42.6%#121 | 80.6%#64 | Vals AI | |
| SWE-bench (Vals)not in index | – | 41.4%#82 | Vals AI | |
| Vibe Code Bench v1.1not in index | – | 1.2%#107 | Vals AI | |
| ALE-Benchnot in index | – | 394.9#109 | Epoch AI Benchmarking Hub |
Agents & tools
| Benchmark | no reasoning | high | thinking | Source | Trend |
|---|---|---|---|---|---|
| Terminal-Bench Hard | 14.4%#189 | – | 24.2%#137 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 63.7%#164 | – | 93.3%#37 | Artificial Analysis | |
| BFCL Overall | 58.3%#10 | – | 69.6%#5 | Berkeley Function Calling Leaderboard | |
| τ³-bench Bankingnot in index | – | 13.1%#23 | – | τ²-bench | |
| ITBench SRE (AA)not in index | 17.9%#39 | – | – | Artificial Analysis | |
| Poker Agentnot in index | – | – | 1079.2%#5 | Vals AI | |
| Terminal-Bench 2.0 (Vals)not in index | 18.0%#57 | – | 24.7%#51 | Vals AI | |
| Vending-Bench 2not in index | – | – | 1106.6#54 | Andon Labs |
Maths
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| ProofBench | – | 4.0%#72 | Vals AI | |
| AIME (Vals) | 27.1%#72 | 91.9%#21 | Vals AI | |
| MGSM | 87.6%#56 | 89.5%#42 | Vals AI | |
| AIME 2026 | – | 94.2%#20 | MathArena | |
| HMMT February 2026 | – | 86.4%#20 | MathArena | |
| MathArena Apexpossible contamination | – | 5.2%#25 | MathArena | |
| Arena Maths | – | 1421#120 | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-Omniscience | -50.9#390 | -29.9#275 | Artificial Analysis | |
| MMLU-Pro | 75.2%#107 | 84.2%#64 | Vals AI | |
| LegalBench | 79.1%#96 | 82.5%#69 | Vals AI | |
| CorpFin | 52.5%#96 | 66.0%#24 | Vals AI | |
| TaxEval | 61.9%#117 | 73.1%#50 | Vals AI | |
| MedQA | 76.0%#75 | 92.1%#34 | Vals AI | |
| CaseLaw v2not in index | – | 60.5%#24 | Vals AI | |
| MedCodenot in index | 28.4%#96 | 28.1%#97 | Vals AI | |
| MedScribenot in index | 77.5%#68 | 78.7%#62 | Vals AI | |
| Public Benefits Benchnot in index | – | 44.8%#46 | Vals AI | |
| AA-Omniscience: accuracynot in index | 17.2%#362 | 25.1%#225 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 17.7%#308 | 26.6%#232 | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| IFBench | 36.5%#292 | 52.7%#153 | Artificial Analysis | |
| Arena Instruction Followingnot in index | – | 1402#135 | Arena (LMArena) |
Human preference
| Benchmark | thinking | Source | Trend |
|---|---|---|---|
| Arena Text | 1430#111 | Arena (LMArena) | |
| Arena Creative Writingnot in index | 1415#86 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1413#135 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1422#128 | Arena (LMArena) |
Multimodal
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| Arena Vision | – | 1201#91 | Arena (LMArena) | |
| MMMU-Pro | 48.4%#244 | 63.3%#190 | Artificial Analysis | |
| MMMU-Pro (Vals)not in index | 63.7%#77 | 72.7%#60 | Vals AI | |
| MortgageTaxnot in index | 34.3%#89 | 42.6%#85 | Vals AI | |
| SAGEnot in index | 11.9%#89 | 31.3%#76 | Vals AI |
Long context
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA-LCR | 31.3%#383 | 74.0%#165 | Artificial Analysis |
Composite
| Benchmark | no reasoning | thinking | Source | Trend |
|---|---|---|---|---|
| AA Intelligence Index v4.3.2 | 11.3#347 | 20.4#225 | Artificial Analysis |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.050 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0002 | $0.0002 | – |
| Summarize a 30-page report | 12,000 / 600 | $0.0027 | $0.0014 | – |
| Code edit | 6,000 / 1,500 | $0.0020 | $0.0013 | – |
| Agentic coding session | 60,000 / 4,000 | $0.014 | $0.0073 | – |
| Structured extraction | 2,000 / 200 | $0.0005 | $0.0003 | – |
More from SpaceXAI
See also
Data as of 21 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Grok 4.1 Fast: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/grok-4-1-fast, data as of 21 Oct 2026.