GPT-5.6 Terra vs Grok 4.7
Full profiles: GPT-5.6 Terra (OpenAI) and Grok 4.7 (SpaceXAI): every benchmark, price and rank.
Updated · Data from 13 sources
As of 6 Nov 2026, GPT-5.6 Terra and Grok 4.7 are neck and neck on quality (60.7 vs 58.3 on the BenchLeader Index, inside the ±5.3 margin), and Grok 4.7 is 1.5× cheaper at $3.00 vs $4.50 per 1M tokens blended. Choose GPT-5.6 Terra for maths (69 vs 56), multimodal (63 vs 56), and reasoning (61 vs 56), where it also streams ~1.5× faster; choose Grok 4.7 for fast first answers (55.1 s vs 96.5 s) and knowledge (61 vs 49) and coding (62 vs 60).
Key facts (as of 6 Nov 2026)
- GPT-5.6 Terra scores 60.7 on the BenchLeader Index and Grok 4.7 scores 58.3 (BenchLeader, 6 Nov 2026).
- GPT-5.6 Terra costs $4.50 per 1M tokens blended (list price, 6 Nov 2026).
- Grok 4.7 costs $3.00 per 1M tokens blended (list price, 6 Nov 2026).
- GPT-5.6 Terra streams ~1.5× faster than Grok 4.7 (Artificial Analysis, 7-day median, 6 Nov 2026).
- Of the 57 benchmarks both models have, GPT-5.6 Terra leads 32 (BenchLeader, 6 Nov 2026).
GPT-5.6 Terra benchmarks 
Grok 4.7 benchmarks 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 | Grok 4.7 | 62 vs 60 category score; FrontierCode 2.0 (AA) 28.7% vs 23.9% | Medium |
| Agents & tool use | GPT-5.6 Terra | 59 vs 57 category score; Terminal-Bench 2.1 (Vals) 77.5% vs 73.4% | Medium |
| Maths | GPT-5.6 Terra | 69 vs 56 category score; FrontierMath Tiers 1–3 86.0% vs 53.0% | High |
| Reasoning | GPT-5.6 Terra | 61 vs 56 category score; GPQA Diamond 93.3% vs 92.7% | Medium |
| Knowledge | Grok 4.7 | 61 vs 49 category score; SimpleQA Verified 56.0% vs 43.2% | High |
| Long documents | GPT-5.6 Terra | 66 vs 63 category score; AA-LCR v1.1 83.0% vs 76.7% | Medium |
| Images and other media | GPT-5.6 Terra | 63 vs 56 category score; SAGE 87.0% vs 88.7% | High |
| Chat quality (human preference) | – | Only Grok 4.7 has an Arena rating (62 category score) | Not comparable |
| Lowest cost | Grok 4.7 | $3.00 vs $4.50 per 1M blended; agentic coding session $0.144 vs $0.168 | – |
| Fastest first answer | Grok 4.7 | 55.1 s (Grok 4.7 (xhigh)) vs 96.5 s (GPT-5.6 Terra (max)) to the first answer token | – |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GPT-5.6 Terra (max) and Grok 4.7 (xhigh) are neck and neck on quality (60.7 vs 58.3, within the ±5.3 margin).
- GPT-5.6 Terra (max) is stronger in agents & tools, instruction following, long context, maths, multimodal, reasoning.
- Grok 4.7 (xhigh) is stronger in coding, knowledge, human preference.
- Grok 4.7 (xhigh) is 1.5× cheaper ($3.00 vs $4.50 per 1M blended).
- GPT-5.6 Terra (max) streams 1.4× faster (100 vs 71 tokens per second).
At a glance
| Spec | GPT-5.6 Terra | Grok 4.7 |
|---|---|---|
| Developer | OpenAI | SpaceXAI |
| Released | 9 Jul 2026 | 21 Sep 2026 |
| Context window (tokens) | 1,050,000 | 500,000 |
| Compared at | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) |
| Open weights | No | No |
| Price in / out / cached input per 1M | $2.00 / $12.00 / $0.200 | $2.00 / $6.00 / $0.500 |
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 | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) | Leader |
|---|---|---|---|
| Reasoning | 61.1 | 56.4 | GPT-5.6 Terra |
| Coding | 59.8 | 61.5 | Grok 4.7 |
| Agents & tools | 58.8 | 56.9 | GPT-5.6 Terra |
| Maths | 68.6 | 55.9 | GPT-5.6 Terra |
| Knowledge | 48.8 | 61.1 | Grok 4.7 |
| Instruction following | 69.9 | – | not comparable |
| Human preference | – | 62.3 | not comparable |
| Multimodal | 63.2 | 55.7 | GPT-5.6 Terra |
| Long context | 66.4 | 63.1 | GPT-5.6 Terra |
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: Instruction following, Human preference.
Where sources disagree
Across the 57 benchmarks both have, GPT-5.6 Terra leads 32, Grok 4.7 leads 23 and 2 are tied. Independent sources split: GPT-5.6 Terra leads on Arena Text; Grok 4.7 leads on AA Intelligence Index v5.0.
| Overall index | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) | Leader |
|---|---|---|---|
| BenchLeader Index | 60.7 | 58.3 | neck and neck |
| AA Intelligence Index v5.0 | 32.3 | 35.5 | Grok 4.7 |
| Arena Text | 1467† | 1441 | GPT-5.6 Terra |
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 | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) |
|---|---|---|
| Blended price per 1M tokens | $4.50 | $3.00 |
| Input / output per 1M | $2.00 / $12.00 | $2.00 / $6.00 |
| Cached input per 1M | $0.200 | $0.500 |
| Price per task: chat reply | $0.0044 | $0.0026 |
| Price per task: summarize a 30-page report | $0.031 | $0.028 |
| Price per task: code edit | $0.030 | $0.021 |
| Price per task: agentic coding session | $0.168 | $0.144 |
| Price per task: structured extraction | $0.0064 | $0.0052 |
| Agentic coding session, input mostly cached | $0.087 | $0.076 |
| Cost per benchmark task (AA Index) | $3.73 | $7.77 |
| Output tokens per task (AA Index) | 66k | 105k |
| Of which reasoning | 41k | 69k |
| Time per task (AA Index) | 11 min | 24 min |
| Output speed | 100 tok/s | 71 tok/s |
| Time to first answer | 96.5 s | 55.1 s |
| Time to a complete answer | 203.7 s | 147.2 s |
Per token is not per task: an average task of Artificial Analysis' Intelligence Index cost $3.73 on GPT-5.6 Terra (max) and $7.77 on Grok 4.7 (xhigh), and Grok 4.7 (xhigh) wrote 1.6× as many output tokens for the same work, which a per-token price hides.
Each model is compared at its best-ranked reasoning-effort setting. GPT-5.6 Terra (max) scores 60.7 and takes 96 s to its first answer token; GPT-5.6 Terra (xhigh) scores 60.3 in 26 s, GPT-5.6 Terra (high) scores 57.0 in 11 s. Grok 4.7 (xhigh) scores 58.3 and takes 55 s to its first answer token; Grok 4.7 (high) scores 57.7 in 31 s, Grok 4.7 (low) scores 56.0 in 9 s. 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
147 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 | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.7 | 58.3 |
| Agents & tools score | BenchLeader | 58.8 | 56.9 |
| Coding score | BenchLeader | 59.8 | 61.5 |
| Instruction following score | BenchLeader | 69.9 | – |
| Knowledge score | BenchLeader | 48.8 | 61.1 |
| Long context score | BenchLeader | 66.4 | 63.1 |
| Maths score | BenchLeader | 68.6 | 55.9 |
| Multimodal score | BenchLeader | 63.2 | 55.7 |
| Reasoning score | BenchLeader | 61.1 | 56.4 |
| Human preference score | BenchLeader | – | 62.3 |
| Blended price $/M | BenchLeader | $4.50 | $3.00 |
| Output speed | BenchLeader | 100 tok/s | 71 tok/s |
| Time to first answer | BenchLeader | 96.5 s | 55.1 s |
| Context window | BenchLeader | 1,050,000 tokens | 500,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 93.3% | 92.7% |
| SimpleBench | SimpleBench | 48.9%† | – |
| Arena Hard Prompts | Arena (LMArena) | 1490† | 1463 |
| LiveBench 2.0 Reasoning | LiveBench | 90.6%† | 82.7%† |
| LiveBench 1.0 Reasoning | LiveBench | 90.6% | 82.7% |
| GPQA Diamond (AA) | Artificial Analysis | 92.5% | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 42.9% | 43.1% |
| HLE-Diamond (AA) | Artificial Analysis | 28.8% | 23.2% |
| GPQA Diamond (Vals) | Vals AI | 90.9%† | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 51.3%† | – |
| ARC-AGI-1 | ARC Prize | 96.5% | 89.5% |
| ARC-AGI-2 | ARC Prize | 83.9% | 61.4% |
| ARC-AGI-3 | ARC Prize | 0.8% | 7.0% |
| CritPt | Artificial Analysis | 30.0% | 17.7% |
| CritPt v2 hard subset (AA) | Artificial Analysis | 50.5% | 33.5% |
| MysteryMechanism | Vals AI | – | 25.2% |
| Chess Puzzles | Epoch AI Benchmarking Hub | 47.0% | 41.0% |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | 35.0% | 29.0% |
| LMCA | Epoch AI Benchmarking Hub | 55.0% | 49.4% |
| DTBench | Epoch AI Benchmarking Hub | 93.3% | 96.0% |
| Coding | |||
| SciCode | SciCode | 55.0% | 57.4% |
| WeirdML | WeirdML | 78.3%† | – |
| FrontierCode | Cognition | 41.3%† | 47.6%† |
| Arena Coding | Arena (LMArena) | 1517† | 1494 |
| Arena WebDev | Arena (LMArena) | 1520† | 1638 |
Show all 147 benchmarks
| Metric | Publisher | GPT-5.6 Terra (max) | Grok 4.7 (xhigh) |
|---|---|---|---|
| Coding | |||
| LiveBench 2.0 Coding | LiveBench | 86.2%† | 84.5%† |
| LiveBench 1.0 Coding | LiveBench | 78.3% | 77.2% |
| LiveCodeBench | Vals AI | 85.9%† | – |
| IOI | Vals AI | 87.6% | 57.7% |
| SWE-bench (Vals) | Vals AI | 95.4% | – |
| Code Migration | Vals AI | 47.8% | 44.8% |
| Code Migration v1.1 | Vals AI | – | 53.2% |
| ProgramBench | Vals AI | 0.5% | 0.5% |
| SWE-bench Multimodal | Vals AI | – | 42.9% |
| Vibe Code Bench 1-100 | Vals AI | 14.8% | – |
| Vibe Code Bench v1.1 | Vals AI | 74.6% | 86.2% |
| Surface Evolver Bench | Epoch AI Benchmarking Hub | 83.8%† | – |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | 69.6% | – |
| CursorBench | Cursor | 41.3% | 46.3% |
| ALE-Bench | Epoch AI Benchmarking Hub | 1951.4 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 29.5% |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | 35.4% | 25.8% |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | 88.0% | – |
| Startrise Frontend | Startrise | 66.9%† | 75.9%† |
| GameBench 2 Overall | DuelLab GameBench 2 | 45.4† | 63.4† |
| GameBench 2 by reasoning setting | DuelLab GameBench 2 | 53.0† | 71.5 |
| FrontierCode 1.1 (AA) | Artificial Analysis | – | 33.8% |
| FrontierCode 2.0 (AA) | Artificial Analysis | 23.9% | 28.7% |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 21.5% | 37.6% |
| APEX-Agents | Mercor | 58.2% | 54.6%† |
| Arena Agent | Arena (LMArena) | -1.5† | 2.7 |
| LiveBench Long Agentic | LiveBench | 33.1%† | 46.1%† |
| LiveBench Browser | LiveBench | 70.3%† | 49.3%† |
| LiveBench Agentic Coding | LiveBench | 55.0% | 54.0% |
| Terminal-Bench Hard | Artificial Analysis | 57.6% | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 86.3% | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | 77.5% | 73.4% |
| GDPval-AA v2.1, normalised score | Artificial Analysis | 47.6% | 60.7% |
| GDPval-AA v2.1 | Artificial Analysis | – | 1718 |
| τ³-Banking (AA) | Artificial Analysis | 40.2% | – |
| ITBench SRE (AA) | Artificial Analysis | 51.0% | 42.1% |
| Analyst Agent (AA) | Artificial Analysis | – | 52.5% |
| APEX-Agents (AA) | Artificial Analysis | 38.9% | – |
| CyberBench | Vals AI | 72.1% | 69.5% |
| Finance Agent v2 | Vals AI | 54.4% | 52.3% |
| Finance Agent v2.1 | Vals AI | – | 60.6% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 18.3% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | 0.8% | 12.5% |
| Legal Research Bench | Vals AI | 41.4% | 47.1% |
| Robo-bench | Vals AI | – | 0.2% |
| SkillsBench | Vals AI | 58.9% | – |
| SREBench | Vals AI | 9.5% | 3.8% |
| Tax Agent Bench | Vals AI | 30.0% | 34.2% |
| Terminal-Bench 4.0 (Vals) | Vals AI | 22.7% | 28.8% |
| Terminal-Bench Science | Vals AI | 10.0% | 10.0% |
| BALROG | BALROG | 53.2% | – |
| Vending-Bench 2 | Andon Labs | 7343.2† | 10536.8† |
| AutomationBench | Artificial Analysis | 59.6% | 65.6% |
| GDP.pdf (AA) | Artificial Analysis | 24.0% | 20.0% |
| Harvey LAB-AA v1.1 | Harvey | – | 9.4% |
| EnterpriseOps-Gym | Artificial Analysis | 38.5% | – |
| AA-Briefcase v1.1 | Artificial Analysis | – | 1637 |
| GDPx | GDPx by SemiAnalysis | – | 3.9% |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | 13.8% | 11.4% |
| CWE-Bench (AA) | Artificial Analysis | – | 68.3% |
| DeepSec-Bench (AA) | Artificial Analysis | – | 26.9% |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 74.0% |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | 86.0% | 53.0% |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | 70.7% | 17.1% |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 99.7% | 98.1% |
| ProofBench | Vals AI | 74.0%† | 26.0%† |
| LiveBench Mathematics | LiveBench | 94.9% | 95.7% |
| Arena Maths | Arena (LMArena) | 1480† | 1443 |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | 43.2% | 56.0% |
| LiveBench 2.0 Data Analysis | LiveBench | 79.3%† | 76.9%† |
| LiveBench Research | LiveBench | 60.4%† | 86.4%† |
| LiveBench 1.0 Data Analysis | LiveBench | 79.3% | 76.9% |
| AA-Omniscience v1.1 | Artificial Analysis | -9.3 | 31.2 |
| MMLU-Pro | Vals AI | 86.7%† | – |
| LegalBench | Vals AI | 85.1% | 84.4% |
| CorpFin | Vals AI | 65.3%† | – |
| TaxEval | Vals AI | 76.2%† | – |
| PRBench Legal | Scale AI SEAL | – | 47.6% |
| BioMysteryBench | Vals AI | – | 69.3% |
| Excel Modeling Benchmark | Vals AI | 66.2% | 67.0% |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 71.5% |
| MedCode | Vals AI | 43.4%† | 49.5% |
| MedScribe | Vals AI | 82.9%† | 89.4% |
| Public Benefits Bench v1.1 | Vals AI | 62.4% | 65.6% |
| GDP.pdf | Epoch AI Benchmarking Hub | 24.7%† | 22.8% |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | 42.3% | 42.7% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | 10.6% | 80.0% |
| HealthBench Professional (AA) | OpenAI | – | 55.8% |
| Instruction following | |||
| LiveBench Language | LiveBench | 82.9% | 80.1% |
| LiveBench Instruction Following | LiveBench | 64.6% | 75.3% |
| IFBench | Artificial Analysis | 71.2% | – |
| Arena Instruction Following | Arena (LMArena) | 1463† | 1438 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1467† | 1441 |
| EQ-Bench 4 | EQ-Bench | 1220† | – |
| Arena Creative Writing | Arena (LMArena) | 1429† | 1419 |
| Arena Multi-turn | Arena (LMArena) | 1474† | 1432 |
| Arena Longer Queries | Arena (LMArena) | 1468† | 1452 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1271† | 1231 |
| LiveBench Design | LiveBench | 76.0%† | 78.0%† |
| MMMU-Pro | Artificial Analysis | 80.7% | 78.2%† |
| MMMU-Pro (Vals) | Vals AI | 86.5% | – |
| MortgageTax | Vals AI | 67.3%† | – |
| SAGE | Vals AI | 87.0% | 88.7% |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | 30.8%† | 32.5%† |
| Code2Video-AA | Artificial Analysis | – | 1127 |
| AA-3D | Artificial Analysis | – | 995 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 83.0% | 76.7% |
| Arena Document | Arena (LMArena) | 1477† | – |
| MLCR | Artificial Analysis | 27.2% | 20.0% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 160.9† | 156.1† |
| LiveBench 2.0 | LiveBench | 67.8%† | 70.1%† |
| LiveBench 1.0 | LiveBench | 77.9% | 77.4% |
| AA Intelligence Index v5.0 | Artificial Analysis | 32.3 | 35.5 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 42.1 | 46.5 |
| AA Progress Index v1 | Artificial Analysis | 323.9 | 322.7 |
| Vals Index | Vals AI | – | 58.6 |
| Vals Finance Index | Vals AI | – | 58.5% |
| Vals RSI Index | Vals AI | – | 20.2% |
| Vals Science Index | Vals AI | – | 34.8% |
| Vals Multimodal Index | Vals AI | 65.1%† | – |
| Vals Router Index | Vals AI | – | 58.6%† |
| AA Cyber Index | Artificial Analysis | – | 56.4 |
Data as of 2026-11-06. Best configuration of each model; every score links to its source on the model pages.
DuelLab GameBench 2, release gb2-20261103-09b2218cbd85 (https://benchmarks.duellab.org/releases/gb2-20261103-09b2218cbd85/), CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Scores mapped to BenchLeader model names; not endorsed by DuelLab.
What changed
- GPT-5.6 Terra (max): BenchLeader Index 61.2 → 60.6.
- GPT-5.6 Terra (max): 3 new results (AA Intelligence Index v5.0, CritPt v2 hard subset (AA), FrontierCode 2.0 (AA)).
- Grok 4.7 (xhigh): 2 score revisions.
- Grok 4.7 (xhigh): 2 score revisions.
- GPT-5.6 Terra (max): 1 new result (MirrorCode).
- Grok 4.7 (xhigh): 1 score revision.
- GPT-5.6 Terra (max): BenchLeader Index 62.5 → 61.2.
- Grok 4.7 (xhigh): BenchLeader Index 59.5 → 58.0.
- GPT-5.6 Terra (max): 2 new results (AA Progress Index v1, HLE-Diamond (AA)); 4 score revisions.
- Grok 4.7 (xhigh): 5 new results (AA Intelligence Index v5.0, AA Progress Index v1, HLE-Diamond (AA)…); 4 score revisions; cost per task moved from $3.24/M to $7.51/M (132% up).
Related comparisons
- GPT-5.6 Terra vs Claude Opus 5.5
- GPT-5.6 Terra vs Gemini 4 Carbon
- GPT-5.6 Terra vs GPT-6 Astra
- Grok 4.7 vs Gemini 4 Argon
- Grok 4.7 vs GPT-6.1 Terra
- Grok 4.7 vs Claude Opus 4.7
- Grok 4.7 vs GPT-5.3-Codex
- GPT-5.6 Terra vs Claude Fable 5.1
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
GPT-5.6 Terra vs Grok 4.7: questions
- Is GPT-5.6 Terra better than Grok 4.7?
- GPT-5.6 Terra (max) and Grok 4.7 (xhigh) are neck and neck on quality (60.7 vs 58.3, within the ±5.3 margin). The BenchLeader Index combines every independent quality benchmark; the two are within the margin of error as of 6 Nov 2026, so check the category scores for your use.
- GPT-5.6 Terra vs Grok 4.7: which is cheaper, and will that change?
- Grok 4.7 is cheaper today: $3.00 against $4.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is GPT-5.6 Terra or Grok 4.7 better for coding?
- Grok 4.7 scores higher in coding (62 vs 60 on the category index, where 50 is average). On FrontierCode 2.0 (AA), the top shared coding benchmark, it is 23.9% for GPT-5.6 Terra against 28.7% for Grok 4.7.
- Which is faster, GPT-5.6 Terra or Grok 4.7?
- GPT-5.6 Terra streams ~1.5× faster (100 vs 71 output tokens per second, Artificial Analysis 7-day median).
- Is GPT-5.6 Terra or Grok 4.7 better for agents and tool use?
- GPT-5.6 Terra scores higher in agentic tasks (59 vs 57 on the category index, where 50 is average). On Terminal-Bench, the top shared agents & tools benchmark, it is 21.5% for GPT-5.6 Terra against 37.6% for Grok 4.7.
More questions (4)
- Which is better for long documents?
- GPT-5.6 Terra scores higher on long-context benchmarks (66 vs 63). GPT-5.6 Terra accepts more context: 1,050,000 against 500,000 tokens for Grok 4.7.
- Which has lower latency?
- Grok 4.7 answers first: 55.1 s against 96.5 s to the first answer token, as Artificial Analysis measures it. Reasoning effort drives most of this: a higher setting thinks longer before it answers.
- Do independent benchmarks agree on GPT-5.6 Terra vs Grok 4.7?
- Across the 57 benchmarks both have, GPT-5.6 Terra leads 32, Grok 4.7 leads 23 and 2 are tied. Independent sources split: GPT-5.6 Terra leads on Arena Text; Grok 4.7 leads on AA Intelligence Index v5.0.
- When was Grok 4.7 released?
- Grok 4.7 was released on 21 Sep 2026.