Gemini 3 Pro vs Step 5 Preview
Full profiles: Gemini 3 Pro (Google) and Step 5 Preview (StepFun): every benchmark, price and rank.
Updated · Data from 15 sources
As of 16 Nov 2026, Gemini 3 Pro is ahead of Step 5 Preview on quality (60.0 vs 56.0 on the BenchLeader Index), but Step 5 Preview is 3.2× cheaper at $1.43 vs $4.50 per 1M tokens blended. Choose Gemini 3 Pro for agents & tools (62 vs 49), reasoning (63 vs 52), and coding (58 vs 50); choose Step 5 Preview for maths (60 vs 54).
Key facts (as of 16 Nov 2026)
- Gemini 3 Pro scores 60.0 on the BenchLeader Index and Step 5 Preview scores 56.0 (BenchLeader, 16 Nov 2026).
- Gemini 3 Pro costs $4.50 per 1M tokens blended (list price, 16 Nov 2026).
- Step 5 Preview costs $1.43 per 1M tokens blended (list price, 16 Nov 2026).
- Of the 10 benchmarks both models have, Gemini 3 Pro leads 9 (BenchLeader, 16 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 | Gemini 3 Pro | 58 vs 50 category score; Arena Coding 1517 vs 1505 | High |
| Agents & tool use | Gemini 3 Pro | 62 vs 49 category score | High |
| Maths | Step 5 Preview | 60 vs 54 category score; ProofBench 42.0% vs 20.0% | Medium |
| Reasoning | Gemini 3 Pro | 63 vs 52 category score; Arena Hard Prompts 1502 vs 1477 | High |
| Knowledge | Gemini 3 Pro | 56 vs 54 category score | Medium |
| Images and other media | Gemini 3 Pro | 66 vs 59 category score; Arena Vision 1305 vs 1263 | Medium |
| Chat quality (human preference) | Gemini 3 Pro | 68 vs 63 on Arena votes | Medium |
| High volume on a budget | Neither flagship | A smaller model is usually the better fit: Cheapest good LLMs, Fastest LLMs | – |
Key differences
- Gemini 3 Pro leads on quality: 60.0 vs 56.0.
- Gemini 3 Pro is stronger in agents & tools, coding, human preference, knowledge, multimodal, reasoning.
- Step 5 Preview is stronger in maths.
- Step 5 Preview is 3.2× cheaper ($1.43 vs $4.50 per 1M blended).
At a glance
| Spec | Gemini 3 Pro | Step 5 Preview |
|---|---|---|
| Developer | StepFun | |
| Released | 18 Nov 2025 | 18 Sep 2026 |
| Context window (tokens) | – | 1,000,000 |
| Compared at | Gemini 3 Pro | Step 5 Preview |
| Open weights | No | Yes |
| Price in / out / cached input per 1M | – | $1.00 / $2.70 / $0.105 |
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 | Step 5 Preview | Leader |
|---|---|---|---|
| Reasoning | 62.6 | 51.7 | Gemini 3 Pro |
| Coding | 57.8 | 49.9 | Gemini 3 Pro |
| Agents & tools | 61.7 | 49.1 | Gemini 3 Pro |
| Maths | 54.0 | 60.1 | Step 5 Preview |
| Knowledge | 55.9 | 53.7 | Gemini 3 Pro |
| Instruction following | 61.1 | – | not comparable |
| Human preference | 67.7 | 63.1 | Gemini 3 Pro |
| Multimodal | 65.6 | 59.3 | Gemini 3 Pro |
| Long context | – | 69.0 | 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: Instruction following, Long context.
Where sources disagree
Across the 10 benchmarks both have, Gemini 3 Pro leads 9, Step 5 Preview leads 1. Every overall index that rates both puts Gemini 3 Pro ahead.
| Overall index | Gemini 3 Pro | Step 5 Preview | Leader |
|---|---|---|---|
| BenchLeader Index | 60.0 | 56.0 | Gemini 3 Pro |
| Arena Text | 1486 | 1449 | 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. Cost, tokens and time per benchmark task are measured: Artificial Analysis’ averages over its whole Intelligence Index run.
| Measure | Gemini 3 Pro | Step 5 Preview |
|---|---|---|
| Blended price per 1M tokens | – | $1.43 |
| Input / output per 1M | – / – | $1.00 / $2.70 |
| Cached input per 1M | – | $0.105 |
| Price per task: chat reply | – | $0.0012 |
| Price per task: summarize a 30-page report | – | $0.014 |
| Price per task: code edit | – | $0.010 |
| Price per task: agentic coding session | – | $0.071 |
| Price per task: structured extraction | – | $0.0025 |
| Agentic coding session, input mostly cached | – | $0.030 |
| Cost per benchmark task (AA Index) | – | $2.35 |
| Output tokens per task (AA Index) | – | 113k |
| Of which reasoning | – | 80k |
| Time per task (AA Index) | – | 15 min |
| Output speed | – | 124 tok/s |
| Time to first answer | – | 23.5 s |
| Time to a complete answer, last measured 13 Nov 2026 | – | 54.4 s |
Each model is compared at its best-ranked reasoning-effort setting. Gemini 3 Pro scores 60.0; Gemini 3 Pro (high) scores 60.0. Step 5 Preview scores 56.0 and takes 24 s to its first answer token. 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
165 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 | Step 5 Preview |
|---|---|---|---|
| BenchLeader Index | BenchLeader | 60.0 | 56.0 |
| Agents & tools score | BenchLeader | 61.7 | 49.1 |
| Coding score | BenchLeader | 57.8 | 49.9 |
| Human preference score | BenchLeader | 67.7 | 63.1 |
| Instruction following score | BenchLeader | 61.1 | – |
| Knowledge score | BenchLeader | 55.9 | 53.7 |
| Maths score | BenchLeader | 54.0 | 60.1 |
| Multimodal score | BenchLeader | 65.6 | 59.3 |
| Reasoning score | BenchLeader | 62.6 | 51.7 |
| Long context score | BenchLeader | – | 69.0 |
| Blended price $/M | BenchLeader | – | $1.43 |
| Output speed | BenchLeader | – | 124 tok/s |
| Time to first answer | BenchLeader | – | 23.5 s |
| Context window | BenchLeader | – | 1,000,000 tokens |
| Reasoning | |||
| GPQA Diamond | Epoch AI Benchmarking Hub | 92.6% | – |
| Humanity's Last Exam | Scale AI / CAIS | 37.5% | – |
| SimpleBench | SimpleBench | 76.4% | – |
| Arena Hard Prompts | Arena (LMArena) | 1502 | 1477 |
| LiveBench 2.0 Reasoning | LiveBench | – | 83.8%† |
| GPQA Diamond (AA) | Artificial Analysis | 90.8%† | – |
| Humanity's Last Exam (AA) | Artificial Analysis | 39.7%† | 46.5% |
| HLE-Diamond (AA) | Artificial Analysis | – | 24.3% |
| 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% | – |
| ARC-AGI-2 | ARC Prize | 31.1% | – |
| CritPt | Artificial Analysis | 9.1%† | 20.9% |
| CritPt v2 hard subset (AA) | Artificial Analysis | – | 32.5% |
| MysteryMechanism | Vals AI | – | 18.5% |
| SciPredict | Scale AI SEAL | 25.3% | – |
| Arena Expert | Arena (LMArena) | 1499 | 1489 |
| CL-bench | Epoch AI Benchmarking Hub | 15.8% | – |
| ForecastBench | ForecastBench | 65.0% | – |
| Coding | |||
| SWE-bench Verified (Epoch) | Epoch AI Benchmarking Hub | 72.9% | – |
| SciCode | SciCode | – | 58.9% |
| WeirdML | WeirdML | 69.9% | – |
| GSO-Bench | GSO-Bench | 18.6% | – |
| Arena Coding | Arena (LMArena) | 1517 | 1505 |
Show all 165 benchmarks
| Metric | Publisher | Gemini 3 Pro | Step 5 Preview |
|---|---|---|---|
| Coding | |||
| Arena WebDev | Arena (LMArena) | 1441 | 1573† |
| LiveBench 2.0 Coding | LiveBench | – | 67.6%† |
| LiveCodeBench | Vals AI | 86.4% | – |
| IOI | Vals AI | – | 38.1% |
| 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 Multilingual (bash only) | SWE-bench | 68.7% | – |
| SWE-bench Verified (any scaffold) | SWE-bench | 77.4% | – |
| Code Migration | Vals AI | – | 39.0% |
| Code Migration v1.1 | Vals AI | – | 47.0% |
| SWE-bench Multimodal | Vals AI | – | 40.8% |
| Vibe Code Bench v1.1 | Vals AI | 14.3%† | 70.4% |
| SWE-Bench Pro (private) | Scale AI SEAL | 17.9% | – |
| ALE-Bench | Epoch AI Benchmarking Hub | 1176.8 | – |
| AlgoTune | AlgoTune | 1.8 | – |
| Terminal-Bench 4.0 (AA) | Artificial Analysis | – | 33.3% |
| Design Arena Website | Design Arena | 1244 | – |
| FrontierCode 1.1 (AA) | Artificial Analysis | – | 38.4% |
| FrontierCode 2.0 (AA) | Artificial Analysis | – | 12.0% |
| LiveCodeBench (AA) | Artificial Analysis | 91.8%† | – |
| Agents & tools | |||
| Terminal-Bench | Terminal-Bench | 69.4% | – |
| GDPval | OpenAI | 40.3% | – |
| Remote Labor Index | Scale AI / CAIS | 1.3% | – |
| Arena Agent | Arena (LMArena) | – | -3.7 |
| LiveBench Long Agentic | LiveBench | – | 19.4%† |
| LiveBench Browser | LiveBench | – | 24.8%† |
| Terminal-Bench Hard | Artificial Analysis | 41.7%† | – |
| τ²-Bench Telecom (AA) | Artificial Analysis | 87.1%† | – |
| MCP Atlas | Scale AI SEAL | 70.3% | – |
| 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 | – | 54.3% |
| GDPval-AA v2.1 | Artificial Analysis | – | 1591 |
| ITBench SRE (AA) | Artificial Analysis | – | 55.6% |
| Analyst Agent (AA) | Artificial Analysis | – | 35.0% |
| APEX-Agents (AA) | Artificial Analysis | – | 38.0% |
| Finance Agent v2 | Vals AI | – | 50.7% |
| Finance Agent v2.1 | Vals AI | – | 58.9% |
| Harvey's Legal Agent Benchmark v1.1 | Vals AI | – | 12.9% |
| Harvey's Legal Agent Benchmark v1 | Vals AI | – | 9.2% |
| Legal Research Bench | Vals AI | – | 40.4% |
| Poker Agent | Vals AI | 1078.9% | – |
| SREBench | Vals AI | – | 0.4% |
| Tax Agent Bench | Vals AI | – | 27.3% |
| Terminal-Bench 2.0 (Vals) | Vals AI | 55.1%† | – |
| Terminal-Bench 4.0 (Vals) | Vals AI | – | 31.8% |
| Terminal-Bench Science | Vals AI | – | 2.9% |
| Arena Search | Arena (LMArena) | 1201 | – |
| Arena Agent Task Completion | Arena (LMArena) | – | -1.5 |
| BALROG | BALROG | 58.1% | – |
| DeepResearch Bench | DeepResearch Bench | 46.3%† | – |
| METR Time Horizons | METR | 71.0% | – |
| Vending-Bench 2 | Andon Labs | 5478.2 | – |
| AutomationBench | Artificial Analysis | – | 51.0% |
| GDP.pdf (AA) | Artificial Analysis | – | 14.8% |
| EnterpriseOps-Gym | Artificial Analysis | – | 47.2% |
| AA-Briefcase v1.1 | Artificial Analysis | – | 1424 |
| Design Arena Agentic | Design Arena | 1080 | – |
| Terminal-Bench-Science 0.1 (AA) | Artificial Analysis | – | 2.4% |
| CWE-Bench (AA) | Artificial Analysis | – | 52.5% |
| DeepSec-Bench (AA) | Artificial Analysis | – | 15.8% |
| CyberGym end-to-end (AA) | Artificial Analysis | – | 74.8% |
| MKA (AA) | Artificial Analysis | – | 27.6% |
| Maths | |||
| OTIS Mock AIME | Epoch AI Benchmarking Hub | 91.4% | – |
| ProofBench | Vals AI | 20.0% | 42.0% |
| 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) | 1475 | 1470 |
| AIME 2025 (AA) | Artificial Analysis | 95.7%† | – |
| Knowledge | |||
| LiveBench 2.0 Data Analysis | LiveBench | – | 63.5%† |
| LiveBench Research | LiveBench | – | 75.5%† |
| AA-Omniscience v1.1 | Artificial Analysis | – | 16.4 |
| 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% | – |
| PRBench Legal | Scale AI SEAL | 40.6% | – |
| MultiNRC | Scale AI SEAL | 59.0% | – |
| MMLU-Pro (HELM) | HELM Capabilities | 90.3% | – |
| BioMysteryBench | Vals AI | – | 67.4% |
| CaseLaw v2 | Vals AI | 53.1%† | – |
| Excel Modeling Benchmark | Vals AI | – | 57.1% |
| Excel Modeling Benchmark v1.1 | Vals AI | – | 69.5% |
| MedCode | Vals AI | 52.2%† | – |
| MedScribe | Vals AI | 72.0%† | – |
| Public Benefits Bench v1.1 | Vals AI | – | 58.1% |
| AA-Omniscience v1.1: accuracy | Artificial Analysis | – | 36.4% |
| AA-Omniscience v1.1: non-hallucination | Artificial Analysis | – | 68.7% |
| Global-MMLU-Lite (AA) | Artificial Analysis | 92.2%† | – |
| Instruction following | |||
| 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 | 1450 |
| Human preference | |||
| Arena Text | Arena (LMArena) | 1486 | 1449 |
| EQ-Bench Creative Writing v3 | EQ-Bench | 1524 | – |
| Arena Creative Writing | Arena (LMArena) | 1484 | 1413 |
| Arena Multi-turn | Arena (LMArena) | 1495 | 1445 |
| Arena Longer Queries | Arena (LMArena) | 1491 | 1461 |
| Multimodal | |||
| Arena Vision | Arena (LMArena) | 1305 | 1263 |
| LiveBench Design | LiveBench | – | 70.8%† |
| MMMU-Pro | Artificial Analysis | 80.2%† | 76.4% |
| VISTA | Scale AI SEAL | 51.5% | – |
| 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% | – |
| Code2Video-AA | Artificial Analysis | – | 821 |
| AA-3D | Artificial Analysis | – | 859 |
| Long context | |||
| AA-LCR v1.1 | Artificial Analysis | 76.0%† | 88.3% |
| Arena Document | Arena (LMArena) | 1451 | – |
| MLCR | Artificial Analysis | – | 18.3% |
| Composite | |||
| Epoch Capabilities Index | Epoch AI Benchmarking Hub | 153.0 | – |
| LiveBench 2.0 | LiveBench | – | 54.9%† |
| AA Intelligence Index v5.0.1 | Artificial Analysis | – | 28.5 |
| AA Intelligence Index v4.3.2 | Artificial Analysis | 28.0† | 43.7 |
| AA Progress Index v1 | Artificial Analysis | 290.7† | 318.2 |
| Vals Index | Vals AI | – | 54.8 |
| HELM Capabilities mean | HELM Capabilities | 79.9% | – |
| Vals Finance Index | Vals AI | – | 55.4% |
| Vals Science Index | Vals AI | – | 29.6% |
| Vals Legal Index | Vals AI | – | 70.2% |
| AA Cyber Index | Artificial Analysis | – | 47.7 |
| 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† | 33.3 |
Data as of 2026-11-16. Best configuration of each model; every score links to its source on the model pages.
What changed
- Step 5 Preview: BenchLeader Index 54.2 → 56.0.
- Gemini 3 Pro: 1 score revision.
- Step 5 Preview: BenchLeader Index 54.7 → 54.2.
- Step 5 Preview: 1 score revision.
- Step 5 Preview: 7 score revisions.
- Gemini 3 Pro: 1 new result (Arena Search).
- Step 5 Preview: 2 new results (Vals Index, Harvey's Legal Agent Benchmark v1.1).
- Step 5 Preview: 1 score revision.
- Step 5 Preview: BenchLeader Index 58.0 → 54.7.
- Step 5 Preview: 5 new results (AA Intelligence Index v5.0.1, AA Progress Index v1, HLE-Diamond (AA)…); 2 score revisions; Step 5 Preview cost per task moved from $1.03/M to $2.35/M (128% up).
Related comparisons
- Gemini 3 Pro vs Gemini 4 Carbon
- Gemini 3 Pro vs Claude Opus 5.5
- Gemini 3 Pro vs GPT-6 Astra
- Step 5 Preview vs Claude Opus 5.5
- Step 5 Preview vs GPT-6 Astra
- Step 5 Preview vs GPT-6.1 Sol
- Step 5 Preview vs Gemini 4 Argon
- Gemini 3 Pro vs GPT-6.1 Sol
Rankings: Best LLM overall · Best LLM for agents · Best cheap LLMs
Gemini 3 Pro vs Step 5 Preview: questions
- Is Gemini 3 Pro better than Step 5 Preview?
- Gemini 3 Pro leads on quality: 60.0 vs 56.0. The BenchLeader Index combines every independent quality benchmark; Gemini 3 Pro is ahead overall as of 16 Nov 2026, but check the category scores for your use.
- Gemini 3 Pro vs Step 5 Preview: which is cheaper, and will that change?
- Step 5 Preview is cheaper today: $1.43 against $4.50 per million tokens, blended at three input tokens per output token. No price change has been announced for either.
- Is Gemini 3 Pro or Step 5 Preview better for coding?
- Gemini 3 Pro scores higher in coding (58 vs 50 on the category index, where 50 is average). On Arena Coding, the top shared coding benchmark, it is 1517 for Gemini 3 Pro against 1505 for Step 5 Preview.
- Is Gemini 3 Pro or Step 5 Preview better for agents and tool use?
- Gemini 3 Pro scores higher in agentic tasks (62 vs 49 on the category index, where 50 is average).
- Do independent benchmarks agree on Gemini 3 Pro vs Step 5 Preview?
- Across the 10 benchmarks both have, Gemini 3 Pro leads 9, Step 5 Preview leads 1. Every overall index that rates both puts Gemini 3 Pro ahead.
More questions (1)
- When was Step 5 Preview released?
- Step 5 Preview was released on 18 Sep 2026.