Gemini 3.1 ProvsGrok 4.5
Gemini 3.1 Pro | Grok 4.5 | |
|---|---|---|
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $2.00 | $2.00 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $12.00 | $6.00 |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.20 | $0.30 |
| Benchmarks | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 37% | 54% |
Gray Swan IPI · k = 1Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. Gray Swan's indirect prompt injection benchmark measures how often such an attack succeeds when the attacker gets a single try. Lower is better. | 14.2% | 13.4% |
Gray Swan IPI · k = 10Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 10 tries and counts an attack as successful if any of them works. Lower is better. | 45.7% | 54.2% |
Gray Swan IPI · k = 15Prompt injection robustness — Attackers hide malicious instructions inside content the AI reads — a web page, an email, a document — and try to hijack what it does. This variant gives the attacker 15 tries and counts an attack as successful if any of them works. Lower is better. | 49.2% | 60.8% |
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | 54.2% | 64.7% |
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | 12% | 54% |
Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better. | 75% | 83% |
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | 70.3% | 83.3% |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | 965 | 1526 |
Arena Elo (Text)Community preference — Real people chat with two anonymous AIs side by side and vote for the answer they prefer. Votes become a chess-style Elo rating on arena.ai — it measures which AI people actually like, not test scores. Higher is better. | 1485 | 1468 |
Arena Elo (Code)Community preference (code) — Like the text arena, but people vote on which AI writes better code. The votes become a chess-style Elo rating on arena.ai. Higher is better. | 1445 | 1555 |
| BenchmarksPublished by one model only | ||
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 80.6% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | — | 78% |
CursorBench v3.2Agentic coding — Cursor's own test of harder, real-world coding tasks inside a code editor, on the refreshed v3.2 task set. Scores aren't comparable with v3.1. Higher is better. | — | 66.7% |
DeepSWE 1.0Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. Higher is better. | — | 62% |
FrontierCode v1.1 (Extended) · extended splitAgentic coding — frontier-difficulty agentic coding tasks (v1.1, extended split) | — | 56.6% |
APEX-SWEExpert software engineering — expert-level software-engineering tasks (AI Productivity Index) | — | 53.6% |
MLE-BenchML engineering — Can the AI do the work of a machine-learning engineer? It competes in real Kaggle competitions — building, training, and tuning models end to end — and the score reflects how well it places. Higher is better. | 42.6% | — |
Frontier-Bench v0.1Agentic computer work — A hard, ever-evolving set of real computer tasks — coding, system administration, data work, and more — that an AI agent has to complete on its own. Run by the Harbor / Laude Institute team as the successor to Terminal-Bench (v0.1 is the first release of the task set). The score is the share of tasks solved. Higher is better. | — | 17.8% |
Terminal-Bench 3.0Agentic terminal coding — command-line task completion (v3.0, much harder task set) | — | 15.7% |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | 68.5% | — |
APEX-AgentsExpert agentic work — expert-level agentic work tasks (AI Productivity Index) | — | 47.1% |
MCP AtlasMulti-step tool use — Can the AI chain together many tools and steps to complete one bigger task, rather than doing just a single thing? Higher is better. | 78.2% | — |
ToolathlonGeneral tool use — Tests how well the AI uses everyday real-world tools and apps to get things done. Higher is better. | 48.8% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 85.9% | — |
Humanity's Last Exam · no toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | 44.4% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 51.4% | — |
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | 77.1% | — |
FrontierMath · Tier 1–3Advanced math — Very hard, research-level math problems. Tiers 1–3 are the (still extremely difficult) lower tiers. Higher is better. | 36.9% | — |
FrontierMath · Tier 4Advanced math — Very hard, research-level math problems. Tier 4 is the hardest — close to what professional research mathematicians tackle. Higher is better. | 16.7% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 94.3% | — |
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | 76.2% | — |
Finance Agent v2Agentic financial analysis — Tests the AI on real financial-analysis work, like digging through reports and making sound decisions. Higher is better. | 43% | — |
Harvey's Legal Agent BenchmarkAgentic legal work — Harvey's test of whether an AI agent can complete real legal work — drafting and reviewing documents, working with spreadsheets and presentations, and navigating files the way a lawyer's assistant would. Higher is better. | — | 12.92% |
MedScribeMedical admin work — Can the AI support doctors with their administrative work, like notes and paperwork? Created by Vals AI. Higher is better. | — | 86.88% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 56 |
GDPval-AAKnowledge work — Measures how well the AI does economically valuable knowledge work, judged against human experts. Shown as a rating (like a chess Elo) — higher is better. | 1314 | — |
AA-BriefcaseKnowledge work — Artificial Analysis agentic office-work eval (Elo) | — | 1313 |
GDPval (win/tie rate)Knowledge work — How often the AI's work matches or beats a human expert's on real knowledge-work tasks. Higher is better. | 67.3% | — |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | 83.3% | — |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | 80.5% | — |
Blueprint-Bench 2Spatial reasoning — Can the AI reason about space and layout — for example, understanding a floor plan or blueprint? Higher is better. | 26.5% | — |
MRCR v2 (8-needle) · 128k averageLong context — Tests whether the AI can find specific details buried inside a very long document (around 128k tokens — roughly a long book). Higher is better. | 84.9% | — |
MRCR v2 (8-needle) · 1M pointwiseLong context — Tests whether the AI can find specific details buried inside an enormous document (around 1 million tokens — many books). Higher is better. | 26.3% | — |
| Overview | ||
| Company | SpaceXAI | |
| Release date | Feb 19 2026 | Jul 8 2026 |
| Access | Proprietary | Proprietary |
Frequently asked questions
Grok 4.5 leads Gemini 3.1 Pro on 8 of the 11 benchmarks they both report. Both charge $2.00 per million input tokens. Grok 4.5 is cheaper on output: $6.00 vs $12.00 per million tokens. Figures are base-tier rates. Gemini 3.1 Pro shipped 139 days before Grok 4.5, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On BullshitBench v2, Grok 4.5 leads at 54% vs Gemini 3.1 Pro at 37%. On Gray Swan IPI · k = 1, Grok 4.5 leads at 13.4% vs Gemini 3.1 Pro at 14.2%. On Gray Swan IPI · k = 10, Gemini 3.1 Pro leads at 45.7% vs Grok 4.5 at 54.2%. On Gray Swan IPI · k = 15, Gemini 3.1 Pro leads at 49.2% vs Grok 4.5 at 60.8%. On SWE-Bench Pro, Grok 4.5 leads at 64.7% vs Gemini 3.1 Pro at 54.2%. On DeepSWE 1.1, Grok 4.5 leads at 54% vs Gemini 3.1 Pro at 12%. On Next.js Evals, Grok 4.5 leads at 83% vs Gemini 3.1 Pro at 75%. On Terminal-Bench 2.1, Grok 4.5 leads at 83.3% vs Gemini 3.1 Pro at 70.3%. On GDPval-AA v2, Grok 4.5 leads at 1526 vs Gemini 3.1 Pro at 965. On Arena Elo (Text), Gemini 3.1 Pro leads at 1485 vs Grok 4.5 at 1468. On Arena Elo (Code), Grok 4.5 leads at 1555 vs Gemini 3.1 Pro at 1445.
Gemini 3.1 Pro was released by Google on Feb 19 2026.
Grok 4.5 was released by SpaceXAI on Jul 8 2026.
Grok 4.5 leads on SWE-Bench Pro — Gemini 3.1 Pro 54.2% vs Grok 4.5 64.7%.
Both charge $2.00 per million input tokens. Grok 4.5 is cheaper on output: $6.00 vs $12.00 per million tokens. Figures are base-tier rates. Rates are pay-as-you-go API prices verified on August 18, 2026.
Other comparisons
Gemini 3.1 ProvsClaude Opus 5Grok 4.5vsClaude Opus 5Gemini 3.1 ProvsGPT-5.6-CyberGrok 4.5vsGPT-5.6-CyberGemini 3.1 ProvsMuse GlimmerGrok 4.5vsMuse GlimmerGemini 3.1 ProvsDeepSeek-V4-Pro-0813Grok 4.5vsDeepSeek-V4-Pro-0813Gemini 3.1 ProvsMistral Medium 3.5Grok 4.5vsMistral Medium 3.5Gemini 3.1 ProvsKimi K3Grok 4.5vsKimi K3