SpaceXAIGrok 4.3
Grok 4.3 is a SpaceXAI proprietary reasoning model, released 17 Apr 2026. Its best configuration (medium reasoning effort) ranks #147 of 785 on the BenchLeader Index at 58.7 ±6.8, in the upper half. The ± is the point: 302 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 instruction following (80) and lowest in reasoning (49). At $1.56 per million tokens blended it is pricier than most ranked models. Output speed of 110 tokens per second puts it faster than most, with a first answer in 12.4 s. It has been measured at 5 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 30 Apr 2026.
- Blended price
- $1.56/M
- $1.25 in · $2.50 out
- Output speed
- 110 tok/s
- measured by Artificial Analysis
- First answer
- 12 s
- first token 1.34 s
- Context
- 1M
- Full answer
- 17 s
- median, reasoning included
- Cost per run
- –
- Released
- 17 Apr 2026
- Overall index59
- Reasoning49
- Agents & tools60
- Knowledge69
- Instruction following80
- Multimodal59
- Long context62
- Composite57
Versions
SpaceXAI has shipped 11 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 4.7high | 21 Sept 2026 | 63.9 | #59 |
| Grok 4.6medium | 11 Aug 2026 | 64.2 | #56 |
| Grok 4.5high | 8 Jul 2026 | 59.3 | #134 |
| Grok 4.3mediumthis page | 17 Apr 2026 | 58.7 | #147 |
| Grok 4.2high | 10 Mar 2026 | – | – |
| Grok 4.20thinking | 17 Feb 2026 | 58.6 | #148 |
| Grok 4.1 | 17 Nov 2025 | 57.2 | #184 |
| Grok 4 | 9 Jul 2025 | 58.0 | #161 |
| Grok 3 | 17 Feb 2025 | 51.5 | #340 |
| Grok 2 | 12 Dec 2024 | 41.9 | #612 |
| Grok 4.20 1 | – | – | – |
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 | 45.9 | #505 | 88 tok/s | 0.66 s | $0.0013 | Agents & tools 41 · Composite 45 · Instruction following 49 · Knowledge 47 · Long context 40 · Multimodal 48 · Reasoning 41 |
| low | 56.9 | #194 | 86 tok/s | 5.96 s | $0.0013 | Agents & tools 57 · Composite 57 · Instruction following 78 · Knowledge 68 · Long context 62 · Multimodal 56 · Reasoning 42 |
| mediumbest | 58.7 | #147 | 110 tok/s | 12 s | $0.0013 | Agents & tools 60 · Composite 57 · Instruction following 80 · Knowledge 69 · Long context 62 · Multimodal 59 · Reasoning 49 |
| high | 57.0 | #188 | 102 tok/s | 17 s | $0.0013 | Agents & tools 45 · Coding 58 · Composite 57 · Instruction following 79 · Knowledge 57 · Long context 61 · Maths 47 · Multimodal 61 · Reasoning 59 |
| not stated | 54.7 | #244 | 80 tok/s | 1.34 s | $0.0013 | Coding 48 · Human preference 52 · Knowledge 60 · Maths 60 · Multimodal 55 · Reasoning 62 |
How its index has moved
8 Sept 2026 to 21 Oct 2026- medium
- high
- low
- not stated
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 | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| GPQA Diamond | – | – | – | 88.8%#57 | – | Epoch AI Benchmarking Hub | |
| Arena Hard Promptseffort not stated | – | – | – | – | 1456#103 | Arena (LMArena) | |
| LiveBench 1.0 Reasoningnot in indexeffort not stated | – | – | – | – | 70.8%#64 | LiveBench | |
| GPQA Diamond (AA)not in index | 65.8%#338 | 84.3%#143 | 89.0%#83 | 90.1%#64 | – | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 6.8%#382 | 18.4%#232 | 30.0%#151 | 37.2%#107 | – | Artificial Analysis | |
| GPQA Diamond (Vals)not in indexeffort not stated | – | – | – | – | 91.4%#22 | Vals AI | |
| CritPt | 0.0%#315 | 0.6%#254 | 4.9%#158 | 8.0%#138 | – | Artificial Analysis | |
| Chess Puzzlesnot in index | – | – | – | 18.0%#99 | – | Epoch AI Benchmarking Hub | |
| LMCAnot in index | – | – | – | 38.3%#114 | – | Epoch AI Benchmarking Hub | |
| DTBenchnot in index | – | – | – | 90.7%#79 | – | Epoch AI Benchmarking Hub | |
| ForecastBenchnot in indexeffort not stated | – | – | – | – | 63.4%#40 | ForecastBench |
Coding
| Benchmark | no reasoning | high | not stated | Source | Trend |
|---|---|---|---|---|---|
| SciCode | – | 48.3%#121 | – | SciCode | |
| WeirdMLeffort not stated | – | – | 49.9%#69 | WeirdML | |
| Arena Codingeffort not stated | – | – | 1488#98 | Arena (LMArena) | |
| Arena WebDeveffort not stated | – | – | 1358#107 | Arena (LMArena) | |
| LiveBench 1.0 Codingnot in indexeffort not stated | – | – | 69.9%#62 | LiveBench | |
| SciCode (AA)not in index | 39.4%#180 | 48.3%#125 | – | Artificial Analysis | |
| LiveCodeBench | – | 84.5%#37 | – | Vals AI | |
| SWE-bench (Vals)not in index | – | 71.4%#55 | – | Vals AI | |
| Code Migrationnot in index | – | 6.8%#65 | – | Vals AI | |
| ProgramBenchnot in index | – | 0.0%#32 | – | Vals AI | |
| Vibe Code Bench v1.1not in index | – | 19.4%#85 | – | Vals AI | |
| ALE-Benchnot in indexeffort not stated | – | – | 944.2#60 | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in index | 0.0%#152 | 0.0%#152 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 34.1%#150 | 39.7%#139 | – | Artificial Analysis |
Agents & tools
| Benchmark | no reasoning | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| LiveBench Agentic Codingnot in indexeffort not stated | – | – | – | – | 18.5%#67 | LiveBench | |
| Terminal-Bench Hard | 18.9%#166 | 26.5%#129 | 30.3%#112 | 37.9%#62 | – | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 65.8%#165 | 88.9%#70 | 91.2%#59 | 97.7%#9 | – | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals) | – | – | – | 42.0%#64 | – | Vals AI | |
| τ³-Banking (AA) | 8.0%#158 | – | – | 12.4%#132 | – | Artificial Analysis | |
| ITBench SRE (AA)not in index | – | – | – | 32.7%#30 | – | Artificial Analysis | |
| Analyst Agent (AA)not in index | – | – | – | 8.8%#45 | – | Artificial Analysis | |
| APEX-Agents (AA)not in index | – | – | – | 17.0%#21 | – | Artificial Analysis | |
| Finance Agent v2not in index | – | – | – | 37.7%#64 | – | Vals AI | |
| Harvey's Legal Agent Benchmark v1not in index | – | – | – | 0.4%#58 | – | Vals AI | |
| Legal Research Benchnot in index | – | – | – | 15.4%#64 | – | Vals AI | |
| SkillsBenchnot in index | – | – | – | 40.6%#28 | – | Vals AI | |
| Tax Agent Benchnot in index | – | – | – | 6.2%#65 | – | Vals AI | |
| Terminal-Bench 2.0 (Vals)not in indexeffort not stated | – | – | – | – | 43.5%#28 | Vals AI | |
| Arena Searchnot in indexeffort not stated | – | – | – | – | 1193#25 | Arena (LMArena) | |
| Vending-Bench 2not in indexeffort not stated | – | – | – | – | 35.3#64 | Andon Labs | |
| AutomationBenchnot in index | 0.9%#198 | – | – | 1.4%#190 | – | Artificial Analysis | |
| GDP.pdf (AA)not in index | 2.8%#183 | – | – | 5.8%#157 | – | Artificial Analysis | |
| EnterpriseOps-Gymnot in index | – | – | – | 26.3%#46 | – | Artificial Analysis |
Maths
| Benchmark | high | not stated | Source | Trend |
|---|---|---|---|---|
| FrontierMath Tiers 1–3 | 42.8%#59 | – | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | 14.6%#61 | – | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 93.3%#59 | – | Epoch AI Benchmarking Hub | |
| ProofBench | 11.0%#64 | – | Vals AI | |
| LiveBench Mathematicsnot in indexeffort not stated | – | 84.3%#56 | LiveBench | |
| Arena Mathseffort not stated | – | 1419#122 | Arena (LMArena) |
Knowledge
| Benchmark | no reasoning | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| SimpleQA Verified | – | – | – | 33.2%#59 | – | Epoch AI Benchmarking Hub | |
| LiveBench 1.0 Data Analysisnot in indexeffort not stated | – | – | – | – | 55.8%#62 | LiveBench | |
| AA-Omniscience | -33.1#292 | 13.9#87 | 16.7#77 | 18.0#76 | – | Artificial Analysis | |
| MMLU-Pro | – | – | – | 85.8%#51 | – | Vals AI | |
| LegalBencheffort not stated | – | – | – | – | 84.5%#32 | Vals AI | |
| CorpFineffort not stated | – | – | – | – | 68.5%#7 | Vals AI | |
| TaxEvaleffort not stated | – | – | – | – | 70.8%#83 | Vals AI | |
| CaseLaw v2not in indexeffort not stated | – | – | – | – | 79.3%#1 | Vals AI | |
| Excel Modeling Benchmarknot in index | – | – | – | 18.4%#68 | – | Vals AI | |
| MedCodenot in indexeffort not stated | – | – | – | – | 38.1%#71 | Vals AI | |
| MedScribenot in indexeffort not stated | – | – | – | – | 74.4%#79 | Vals AI | |
| Public Benefits Bench v1.1not in index | – | – | – | 51.7%#43 | – | Vals AI | |
| GDP.pdfnot in index | – | – | – | 8.0%#53 | 8.0%#53 | Epoch AI Benchmarking Hub | |
| AA-Omniscience: accuracynot in index | 23.7%#251 | 26.7%#213 | 28.8%#194 | 34.8%#157 | – | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in index | 25.5%#241 | 82.6%#9 | 83.1%#7 | 74.2%#27 | – | Artificial Analysis |
Instruction following
| Benchmark | no reasoning | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| LiveBench Languagenot in indexeffort not stated | – | – | – | – | 73.6%#55 | LiveBench | |
| LiveBench Instruction Followingnot in indexeffort not stated | – | – | – | – | 62.8%#52 | LiveBench | |
| IFBench | 47.6%#187 | 81.0%#6 | 83.3%#1 | 81.3%#4 | – | Artificial Analysis | |
| Arena Instruction Followingnot in indexeffort not stated | – | – | – | – | 1415#117 | Arena (LMArena) |
Human preference
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Arena Text | 1442#90 | Arena (LMArena) | |
| EQ-Bench 4 | 1075#24 | EQ-Bench | |
| Arena Creative Writingnot in index | 1422#79 | Arena (LMArena) | |
| Arena Multi-turnnot in index | 1447#93 | Arena (LMArena) | |
| Arena Longer Queriesnot in index | 1443#99 | Arena (LMArena) |
Multimodal
| Benchmark | no reasoning | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| Arena Visioneffort not stated | – | – | – | – | 1229#78 | Arena (LMArena) | |
| MMMU-Pro | 64.8%#181 | 72.8%#129 | 75.8%#95 | 78.1%#77 | – | Artificial Analysis | |
| MMMU-Pro (Vals)not in indexeffort not stated | – | – | – | – | 83.1%#34 | Vals AI | |
| MortgageTaxnot in indexeffort not stated | – | – | – | – | 48.3%#81 | Vals AI | |
| SAGEnot in indexeffort not stated | – | – | – | – | 19.7%#87 | Vals AI | |
| Blueprint-Bench 2not in indexeffort not stated | – | – | – | – | 0.0%#29 | Epoch AI Benchmarking Hub |
Long context
| Benchmark | no reasoning | low | medium | high | Source | Trend |
|---|---|---|---|---|---|---|
| AA-LCR | 32.3%#384 | 74.0%#168 | 75.0%#157 | 73.0%#182 | Artificial Analysis | |
| MLCRnot in index | 5.6%#139 | – | – | 8.3%#127 | Artificial Analysis |
Composite
| Benchmark | no reasoning | low | medium | high | not stated | Source | Trend |
|---|---|---|---|---|---|---|---|
| Epoch Capabilities Indexnot in indexeffort not stated | – | – | – | – | 148.8#64 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0not in indexeffort not stated | – | – | – | – | 62.3%#66 | LiveBench | |
| AA Intelligence Index v4.3.2 | 14.0#302 | 24.3#176 | 24.8#169 | 24.9#167 | – | Artificial Analysis | |
| Vals Multimodal Indexnot in index | – | – | – | 43.3%#27 | – | Vals AI |
Where it wins
Benchmarks where this configuration ranks in the top five of every configuration measured.
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| SpaceXAI | 80 tok/s | 1.34 s | $1.25 | $2.50 | 1M | – |
| SpaceXAI | 77 tok/s | 1.22 s | $1.25 | $2.50 | 1M | – |
| SpaceXAI | 63 tok/s | 0.57 s | $2.50 | $5.00 | 1M | – |
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 $0.200 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0013 | $0.0009 | 15.2 s |
| Summarize a 30-page report | 12,000 / 600 | $0.017 | $0.0071 | 17.9 s |
| Code edit | 6,000 / 1,500 | $0.011 | $0.0065 | 26.0 s |
| Agentic coding session | 60,000 / 4,000 | $0.085 | $0.038 | 48.7 s |
| Structured extraction | 2,000 / 200 | $0.0030 | $0.0014 | 14.3 s |
More from SpaceXAI
See also
Questions people ask
- What are Grok 4.3's benchmark scores?
- As of 24 Oct 2026, Grok 4.3 (medium effort) scores 58.7 ±6.8 on the BenchLeader Index, strongest in instruction following (80) and weakest in reasoning (49). That rests on 7 independent benchmark results, and every score on this page links to its publisher.
- How does Grok 4.3 rank?
- Grok 4.3 (medium effort) is #147 of 785 ranked configurations on 24 Oct 2026. 302 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Grok 4.3 cost per 1M tokens?
- As of 24 Oct 2026, Grok 4.3 costs $1.25 per 1M input tokens and $2.50 per 1M output tokens, $1.56 blended (three input tokens per output token).
- When was Grok 4.3 released?
- SpaceXAI released Grok 4.3 on 17 Apr 2026. BenchLeader has results for it at 5 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Grok 4.3: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/grok-4-3.