Gemini 3 Pro vs GPT-6 Astra
Full profiles: Gemini 3 Pro (Google) and GPT-6 Astra (OpenAI): every benchmark, price and rank.
Updated · Data from 16 sources
As of 29 Oct 2026, GPT-6 Astra is ahead of Gemini 3 Pro on quality (68.2 vs 60.2 on the BenchLeader Index). GPT-6 Astra leads in every category we measure both in.
Key facts (as of 29 Oct 2026)
- Gemini 3 Pro scores 60.2 on the BenchLeader Index and GPT-6 Astra scores 68.2 (BenchLeader, 29 Oct 2026).
- GPT-6 Astra costs $20.00 per 1M tokens blended (list price, 29 Oct 2026).
- Of the 8 benchmarks both models have, GPT-6 Astra leads 8 (BenchLeader, 29 Oct 2026).
Gemini 3 Pro benchmarks 
GPT-6 Astra 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 | GPT-6 Astra | 83 vs 58 category score; GSO-Bench 79.4% vs 18.6% | High |
| Agents & tool use | GPT-6 Astra | 70 vs 62 category score; Remote Labor Index 20.8% vs 1.3% | High |
| Maths | GPT-6 Astra | 78 vs 54 category score; ProofBench 99.0% vs 20.0% | High |
| Reasoning | GPT-6 Astra | 86 vs 66 category score; Humanity's Last Exam 54.8% vs 37.5% | High |
| Knowledge | Either | 56 vs 56 category score; PRBench Finance 39.2% vs 47.5% | Either |
| Chat quality (human preference) | – | Only Gemini 3 Pro has an Arena rating (68 category score) | Not comparable |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- GPT-6 Astra leads on quality: 68.2 vs 60.2.
- Gemini 3 Pro is stronger in human preference, instruction following, knowledge, multimodal.
- GPT-6 Astra is stronger in agents & tools, coding, maths, reasoning, composite.
At a glance
| Spec | Gemini 3 Pro | GPT-6 Astra |
|---|---|---|
| Developer | OpenAI | |
| Released | 18 Nov 2025 | 2 Sept 2026 |
| Context window (tokens) | – | 1,050,000 |
| Compared at | Gemini 3 Pro | GPT-6 Astra |
| Open weights | No | No |
| Price in / out / cached input per 1M | – | $10.00 / $50.00 / $1.00 |
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 | Gemini 3 Pro | GPT-6 Astra | Leader |
|---|---|---|---|
| Reasoning | 65.6 | 85.6 | GPT-6 Astra |
| Coding | 57.8 | 83.2 | GPT-6 Astra |
| Agents & tools | 61.6 | 69.6 | GPT-6 Astra |
| Maths | 54.2 | 77.5 | GPT-6 Astra |
| Knowledge | 55.9 | 55.7 | level |
| Instruction following | 59.9 | – | not comparable |
| Human preference | 67.8 | – | not comparable |
| Multimodal | 65.6 | – | not comparable |
| Composite | – | 73.7 | not comparable |
Where sources disagree
Across the 8 benchmarks both have, Gemini 3 Pro leads 0, GPT-6 Astra leads 8. Independent sources split: Gemini 3 Pro leads on Arena Text; GPT-6 Astra leads on BenchLeader Index.
| Overall index | Gemini 3 Pro | GPT-6 Astra | Leader |
|---|---|---|---|
| BenchLeader Index | 60.2 | 68.2 | GPT-6 Astra |
| Arena Text | 1486 | 1477† | Gemini 3 Pro |
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; suite cost is what Artificial Analysis paid to run its whole Intelligence Index.
| Measure | Gemini 3 Pro | GPT-6 Astra |
|---|---|---|
| Blended price per 1M tokens | – | $20.00 |
| Input / output per 1M | – / – | $10.00 / $50.00 |
| Cached input per 1M | – | $1.00 |
| Price per task: chat reply | – | $0.019 |
| Price per task: summarize a 30-page report | – | $0.150 |
| Price per task: code edit | – | $0.135 |
| Price per task: agentic coding session | – | $0.800 |
| Price per task: structured extraction | – | $0.030 |
| Output speed | – | 40 tok/s |
| Time to first answer | – | 3.0 s |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3 Pro scores 60.2; Gemini 3 Pro (high) scores 60.1. GPT-6 Astra scores 68.2 and takes 3 s to its first answer token; GPT-6 Astra (max) scores 67.7 in 232 s, GPT-6 Astra (high) scores 66.8 in 33 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
199 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 | Gemini 3 Pro | GPT-6 Astra |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.2 | 68.2 |
| Agents & tools score | BenchLeader | 61.6 | 69.6 |
| Coding score | BenchLeader | 57.8 | 83.2 |
| Human preference score | BenchLeader | 67.8 | – |
| Instruction following score | BenchLeader | 59.9 | – |
| Knowledge score | BenchLeader | 55.9 | 55.7 |
| Maths score | BenchLeader | 54.2 | 77.5 |
| Multimodal score | BenchLeader | 65.6 | – |
| Reasoning score | BenchLeader | 65.6 | 85.6 |
| Composite score | BenchLeader | – | 73.7 |
| Blended price $/M | BenchLeader | – | $20.00 |
| Output speed | BenchLeader | – | 40 tok/s |
| Time to first answer | BenchLeader | – | 3.0 s |
| Context window | BenchLeader | – | 1,050,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 92.6% | 95.8%† |
| Humanity's Last Exam | Scale AI / CAIS | 37.5% | 54.8% |
| SimpleBench | SimpleBench | 76.4% | – |
| Arena Hard Prompts | Arena (LMArena) | 1502 | 1501† |
| LiveBench 2.0 Reasoning | LiveBench | – | 92.7% |
| LiveBench 1.0 Reasoning | LiveBench | – | 92.7%† |
| GPQA Diamond (AA) | Artificial Analysis | 90.8%† | 96.1%† |
| Humanity's Last Exam (AA) | Artificial Analysis | 39.7%† | 54.7%† |
| HLE-Diamond (AA) | Artificial Analysis | – | 63.8%† |
| GPQA Diamond (Vals) | Vals AI | 91.7%† | – |
| Kagi LLM Benchmark | Kagi LLM Benchmark | 80.1% | – |
| GPQA Diamond (HELM) | HELM Capabilities | 80.3% | – |
| ARC-AGI-1 | ARC Prize | 75.0% | 97.5%† |
| ARC-AGI-2 | ARC Prize | 54.0% | 95.0%† |
| ARC-AGI-3 | ARC Prize | – | 98.6%† |
| CritPt | Artificial Analysis | 9.1%† | 31.7%† |
| CritPt v2 hard subset (AA) | Artificial Analysis | – | 70.5%† |
| MysteryMechanism | Vals AI | – | 53.1%† |
| SciPredict | Scale AI SEAL | 25.3% | – |
| Chess Puzzles | Epoch AI Benchmarking Hub | – | 74.0%† |
| Mystery Game Puzzles | Epoch AI Benchmarking Hub | – | 84.0%† |
| CL-bench | Epoch AI Benchmarking Hub | 15.8% | – |
| LMCA | Epoch AI Benchmarking Hub | – | 64.1% |
| DTBench | Epoch AI Benchmarking Hub | – | 97.1% |
| ForecastBench | ForecastBench | 65.2% | – |
Show all 199 benchmarks
| Metric | Publisher | Gemini 3 Pro | GPT-6 Astra |
|---|---|---|---|
| Reasoning | |||
| BTF-3 | Epoch AI Benchmarking Hub | – | 13.5%† |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 72.9% | – |
| SciCode | SciCode | – | 56.5%† |
| WeirdML | WeirdML | 69.9% | 93.3%† |
| FrontierCode | Cognition | – | 53.3%† |
| GSO-Bench | GSO-Bench | 18.6% | 79.4% |
| Arena Coding | Arena (LMArena) | 1517 | 1543† |
| Arena WebDev | Arena (LMArena) | 1440 | 1774† |
| LiveBench 2.0 Coding | LiveBench | – | 93.7% |
| LiveBench 1.0 Coding | LiveBench | – | 80.4%† |
| LiveCodeBench | Vals AI | 86.4% | – |
| IOI | Vals AI | – | 100.0%† |
| SWE-bench (Vals) | Vals AI | 76.4%† | – |
| SWE-Bench Pro | Scale AI SEAL | 43.3% | – |
| SWE-bench Verified (bash only) | SWE-bench | 74.2% | – |
| SWE-bench Verified (any scaffold) | SWE-bench | 77.4% | – |
| Code Migration | Vals AI | – | 67.7%† |
| Code Migration v1.1 | Vals AI | – | 80.6%† |
| ProgramBench | Vals AI | – | 5.5%† |
| SWE-bench Multimodal | Vals AI | – | 43.3%† |
| Vibe Code Bench 1-100 | Vals AI | – | 27.6%† |
| Vibe Code Bench v1.1 | Vals AI | 14.3%† | 89.6%† |
| SWE-Bench Pro (private) | Scale AI SEAL | 43.3% | – |
| SWE Atlas: Codebase QnA | Scale AI SEAL | – | 59.1%† |
| SWE Atlas: Refactoring | Scale AI SEAL | – | 59.0%† |
| SWE Atlas: Test Writing | Scale AI SEAL | – | 50.7%† |
| MirrorCode | Epoch AI Benchmarking Hub | – | 46.7%† |
| DeepSWE v1.1 | Epoch AI Benchmarking Hub | – | 73.2%† |
| ALE-Bench | Epoch AI Benchmarking Hub | 1176.8 | 2951.3† |
| AlgoTune | AlgoTune | 1.8 | – |
| FrontierSWE | Epoch AI Benchmarking Hub | – | 65.5%† |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | – | 59.1%† |
| Terminal-Bench 2.1 (AA) | Artificial Analysis | – | 88.4%† |
| Startrise Frontend | Startrise | – | 79.9% |
| FrontierCode 1.1 (AA) | Artificial Analysis | – | 47.0%† |
| FrontierCode 2.0 (AA) | Artificial Analysis | – | 42.4%† |
| LiveCodeBench (AA) | Artificial Analysis | 91.8%† | – |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 69.4% | 58.2%† |
| GDPval | OpenAI | 40.3% | – |
| Remote Labor Index | Scale AI / CAIS | 1.3% | 20.8% |
| APEX-Agents | Mercor | – | 64.7% |
| Arena Agent | Arena (LMArena) | – | 14† |
| LiveBench Long Agentic | LiveBench | – | 77.5% |
| LiveBench Browser | LiveBench | – | 91.3% |
| LiveBench Agentic Coding | LiveBench | – | 57.3%† |
| Terminal-Bench Hard | Artificial Analysis | 41.7%† | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 87.1%† | – |
| Terminal-Bench 2.1 (Vals) | Vals AI | – | 87.3%† |
| MCP Atlas | Scale AI SEAL | 70.3% | – |
| HiL-Bench | Scale AI SEAL | – | 65.3% |
| BFCL Overall | Berkeley Function Calling Leaderboard | 72.5% | – |
| τ²-bench | τ²-bench | 82.5%† | – |
| τ³-bench Banking | τ²-bench | 18.0%† | – |
| GDPval-AA v2.1, normalised score | Artificial Analysis | – | 53.7%† |
| GDPval-AA v2.1 | Artificial Analysis | – | 1584† |
| τ³-Banking (AA) | Artificial Analysis | – | 41.4%† |
| ITBench SRE (AA) | Artificial Analysis | – | 48.6%† |
| Analyst Agent (AA) | Artificial Analysis | – | 51.3%† |
| CUA-bench | Vals AI | – | 19.2%† |
| CyberBench | Vals AI | – | 41.1%† |
| Finance Agent v2 | Vals AI | – | 53.5%† |
| Finance Agent v2.1 | Vals AI | – | 60.4%† |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 7.1%† |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 5.4%† |
| Legal Research Bench | Vals AI | – | 39.4%† |
| Poker Agent | Vals AI | 1078.9% | – |
| Robo-bench | Vals AI | – | 23.3%† |
| SREBench | Vals AI | – | 56.9%† |
| Tax Agent Bench | Vals AI | – | 20.7%† |
| Terminal-Bench 2.0 (Vals) | Vals AI | 55.1%† | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 59.6%† |
| Terminal-Bench Science | Vals AI | – | 62.9%† |
| Time Horizon Index: KSP | Vals AI | – | 90.5%† |
| Arena Search | Arena (LMArena) | – | 1246† |
| BALROG | BALROG | 58.1% | 68.3%† |
| DeepResearch Bench | DeepResearch Bench | 46.3%† | – |
| METR Time Horizons | METR | 71.0% | – |
| Vending-Bench 2 | Andon Labs | 5478.2 | 15514.7 |
| AutomationBench | Artificial Analysis | – | 68.5%† |
| GDP.pdf (AA) | Artificial Analysis | – | 31.0%† |
| Harvey LAB-AA v1.1 | Harvey | – | 8.6%† |
| AA-Briefcase v1.1 | Artificial Analysis | – | 1568† |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | – | 63.3%† |
| CWE-Bench (AA) | Artificial Analysis | – | 63.3%† |
| DeepSec-Bench (AA) | Artificial Analysis | – | 36.9%† |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 0.0%† |
| FACTS Search | – | 89.9% | |
| Maths | |||
| FrontierMath Tiers 1–3 | Epoch AI Benchmarking Hub | – | 93.7%† |
| FrontierMath Tier 4 | Epoch AI Benchmarking Hub | – | 97.6%† |
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 91.4% | 100.0%† |
| ProofBench | Vals AI | 20.0% | 99.0% |
| LiveBench Mathematics | LiveBench | – | 96.8%† |
| AIME (Vals) | Vals AI | 96.7%† | – |
| MGSM | Vals AI | 93.9%† | – |
| AIME 2026 | MathArena | 91.7% | – |
| HMMT February 2026 | MathArena | 86.4% | – |
| MathArena Apex | MathArena | 23.4% | – |
| Omni-MATH (HELM) | HELM Capabilities | 55.6% | – |
| MATH 500 | Vals AI | 96.4%† | – |
| Arena Maths | Arena (LMArena) | 1477 | 1487† |
| Knowledge | |||
| SimpleQA Verified | Epoch AI Benchmarking Hub | – | 75.6%† |
| LiveBench 2.0 Data Analysis | LiveBench | – | 83.0% |
| LiveBench Research | LiveBench | – | 85.2% |
| LiveBench 1.0 Data Analysis | LiveBench | – | 83.0%† |
| AA-Omniscience v1.1 | Artificial Analysis | – | 42.4† |
| MMLU-Pro | Vals AI | 90.1% | – |
| LegalBench | Vals AI | 87.0%† | – |
| CorpFin | Vals AI | 63.7%† | – |
| TaxEval | Vals AI | 72.6%† | – |
| MedQA | Vals AI | 96.0%† | – |
| PRBench Finance | Scale AI SEAL | 39.2% | 47.5% |
| PRBench Legal | Scale AI SEAL | 40.6% | 48.4% |
| MultiNRC | Scale AI SEAL | 59.0% | – |
| MMLU-Pro (HELM) | HELM Capabilities | 90.3% | – |
| BioMysteryBench | Vals AI | – | 79.3%† |
| CaseLaw v2 | Vals AI | 53.1%† | – |
| Excel Modeling Benchmark | Vals AI | – | 71.7%† |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 72.7%† |
| MedCode | Vals AI | 52.2%† | 48.5%† |
| MedScribe | Vals AI | 72.0%† | 87.9%† |
| DrugDiscoveryBench | Scale AI SEAL | – | 68.7%† |
| EBR-bench | Epoch AI Benchmarking Hub | – | 76.2%† |
| GDP.pdf | Epoch AI Benchmarking Hub | – | 34.2%† |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | – | 61.6%† |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | – | 50.2%† |
| Global-MMLU-Lite (AA) | Artificial Analysis | 92.2%† | – |
| HealthBench Professional (AA) | OpenAI | – | 66.0%† |
| Instruction following | |||
| LiveBench Language | LiveBench | – | 89.4%† |
| LiveBench Instruction Following | LiveBench | – | 75.6%† |
| IFBench | Artificial Analysis | 70.4%† | – |
| MultiChallenge | Scale AI SEAL | 65.7% | – |
| TutorBench | Scale AI SEAL | 53.7% | – |
| IFEval (HELM) | HELM Capabilities | 87.6% | – |
| WildBench (HELM) | HELM Capabilities | 85.9% | – |
| Arena Instruction Following | Arena (LMArena) | 1473 | 1473† |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1486 | 1477† |
| Arena Creative Writing | Arena (LMArena) | 1485 | 1450† |
| Arena Multi-turn | Arena (LMArena) | 1496 | 1485† |
| Arena Longer Queries | Arena (LMArena) | 1491 | 1489† |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1305 | 1280† |
| LiveBench Design | LiveBench | – | 82.0% |
| MMMU-Pro | Artificial Analysis | 80.2%† | 86.9%† |
| VISTA | Scale AI SEAL | 51.5% | – |
| MMMU-Pro (official) | MMMU | 81.0% | – |
| MMMU-Pro (Vals) | Vals AI | 87.5%† | – |
| MortgageTax | Vals AI | 69.1%† | – |
| AudioMC | Scale AI SEAL | 54.6%† | – |
| AudioMC (text output) | Scale AI SEAL | 54.6%† | – |
| VTB | Scale AI SEAL | 26.9% | – |
| GeoBench | GeoBench | 84.0% | – |
| VPCT | Epoch AI Benchmarking Hub | 91.0% | – |
| Blueprint-Bench 2 | Epoch AI Benchmarking Hub | – | 49.7% |
| Code2Video-AA | Artificial Analysis | – | 1427† |
| AA-3D | Artificial Analysis | – | 1164† |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 76.0%† | 80.7%† |
| Arena Document | Arena (LMArena) | 1451 | 1473† |
| MLCR | Artificial Analysis | – | 41.7%† |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 153.1 | 168.8 |
| LiveBench 2.0 | LiveBench | – | 85.6% |
| LiveBench 1.0 | LiveBench | – | 82.2%† |
| AA Intelligence Index v5.0 | Artificial Analysis | – | 56.9† |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 28.0† | 52.7† |
| AA Progress Index v1 | Artificial Analysis | 290.7† | 349.6† |
| Vals Index | Vals AI | – | 65.0† |
| HELM Capabilities mean | HELM Capabilities | 79.9% | – |
| Vals Finance Index | Vals AI | – | 55.6%† |
| Vals RSI Index | Vals AI | – | 27.1%† |
| Vals Science Index | Vals AI | – | 65.1%† |
| Vals Router Index | Vals AI | – | 65.0% |
| AA Cyber Index | Artificial Analysis | – | 33.4† |
| Safety & honesty | |||
| FORTRESS | Scale AI SEAL | 41.7% | – |
| MASK | Scale AI SEAL | 42.6% | – |
| PropensityBench | Scale AI SEAL | 52.9% | – |
| Metadata | |||
| AA Openness Index | Artificial Analysis | 5.6† | – |
Data as of 2026-10-29. Best configuration of each model; every score links to its source on the model pages.
What changed
- GPT-6 Astra: 3 score revisions.
- Gemini 3 Pro: 1 score revision.
- GPT-6 Astra: 1 new result (Vals Router Index).
- GPT-6 Astra: 1 score revision.
- Gemini 3 Pro: 1 score revision.
- GPT-6 Astra: 8 new results (LiveBench 2.0, LiveBench Long Agentic, LiveBench 2.0 Reasoning…).
- GPT-6 Astra: 1 new result (Startrise Frontend).
- Gemini 3 Pro: BenchLeader Index 60.9 → 60.4.
- GPT-6 Astra: BenchLeader Index 68.9 → 68.0.
- GPT-6 Astra: BenchLeader Index 67.3 → 68.9.
Related comparisons
- Gemini 3 Pro vs Claude Opus 5.5
- Gemini 3 Pro vs GPT-6.1 Sol
- Gemini 3 Pro vs Claude Fable 5.1
- GPT-6 Astra vs Claude Opus 5.5
- GPT-6 Astra vs GPT-6.1 Sol
- GPT-6 Astra vs Claude Fable 5.1
- GPT-6 Astra vs Claude Fable 5
- Gemini 3 Pro vs GPT-5.6 Sol
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 3 Pro vs GPT-6 Astra: questions
- Is GPT-6 Astra better than Gemini 3 Pro?
- GPT-6 Astra leads on quality: 68.2 vs 60.2. The BenchLeader Index combines every independent quality benchmark; GPT-6 Astra is ahead overall as of 29 Oct 2026, but check the category scores for your use.
- Is Gemini 3 Pro or GPT-6 Astra better for coding?
- GPT-6 Astra scores higher in coding (83 vs 58 on the category index, where 50 is average). On GSO-Bench, the top shared coding benchmark, it is 18.6% for Gemini 3 Pro against 79.4% for GPT-6 Astra.
- Is Gemini 3 Pro or GPT-6 Astra better for agents and tool use?
- GPT-6 Astra scores higher in agentic tasks (70 vs 62 on the category index, where 50 is average). On Remote Labor Index, the top shared agents & tools benchmark, it is 1.3% for Gemini 3 Pro against 20.8% for GPT-6 Astra.
- Do independent benchmarks agree on Gemini 3 Pro vs GPT-6 Astra?
- Across the 8 benchmarks both have, Gemini 3 Pro leads 0, GPT-6 Astra leads 8. Independent sources split: Gemini 3 Pro leads on Arena Text; GPT-6 Astra leads on BenchLeader Index.
- When was GPT-6 Astra released?
- GPT-6 Astra was released on 2 Sept 2026.