Full comparison
The complete AI subscription matrix
All 70+ parameters across 8 weighted categories, plus technical specs and limits. Scroll sideways on mobile.
Legend: ●●●● = best, fewer dots = weaker (quality ratings). ✓ yes · ~ partial/limited · ✗ no. Bold rows are the computed category score (0–100). Ratings reflect general standing as of 14 June 2026; specs and prices are approximate – verify officially.
| Feature | ChatGPT | Claude | Gemini | Perplexity | Grok | Copilot | MultipleChat |
|---|---|---|---|---|---|---|---|
| Technical specs & limits | info | info | info | info | info | info | info |
| Flagship model | Latest GPT | Latest Claude | Latest Gemini | Several (routed) | Latest Grok | Latest GPT (OpenAI) | ChatGPT, Claude, Gemini, Grok |
| Context window | 128K tokens | 200K tokens | up to 1M tokens | ~128K tokens | 131K tokens | 128K tokens | up to 200K tokens |
| Max output (approx.) | ~16K tokens | ~32–64K tokens | ~8–64K tokens | ~4–8K tokens | ~16K tokens | ~16K tokens | per model |
| Standard usage limit (approx.) | ~80 msgs / 3h | ~45 msgs / 5h | generous daily cap | 300+ Pro searches / day | high daily cap | priority, fair-use | per-model, generous |
| Max file upload (approx.) | ~512 MB | ~30 MB | ~100 MB | ~50 MB | ~20 MB | ~30 MB | up to 200 MB (higher tiers) |
| Files per chat (approx.) | ~10–20 | ~20 | varies | many | few | varies | varies |
| Supported file types | PDF, Office, CSV, images, code | PDF, text, CSV, images | PDF, Office, Drive, images | PDF, CSV, images, docs | images, some docs | Office, PDF, images | All file types (PDF, Office, CSV, images, code + Mistral OCR) |
| Image generation limit | daily cap | — | daily cap | limited (Pro) | daily cap | daily boosts | 8 models, per-plan |
| Live web access | yes | limited | yes (Google) | yes (core) | yes (X + web) | yes (Bing) | yes (Perplexity) |
| Languages | 95+ | 95+ | 100+ | many | many | 95+ | many |
| Intelligence & model quality · 30% | 95 | 82 | 88 | 75 | 61 | 75 | 88 |
| General reasoning (GPQA / MMLU-Pro) | ●●●● | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●● |
| Math & logic (AIME / MATH) | ●●●● | ●●●○ | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●● |
| Coding (SWE-bench / LiveCodeBench) | ●●●● | ●●●● | ●●●○ | ●●○○ | ●○○○ | ●●●○ | ●●●● |
| Deep research mode | ●●●● | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●● |
| Agentic tasks / tool use | ●●●● | ●●○○ | ●●●○ | ●●○○ | ●●○○ | ●●○○ | ●●○○ |
| Long-context handling | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●● |
| Multimodal understanding | ●●●● | ●●●○ | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●○ |
| Multilingual ability (MMMLU) | ●●●● | ●●●○ | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●○ |
| Instruction following / steerability | ●●●● | ●●●● | ●●●○ | ●●●○ | ●○○○ | ●●●○ | ●●●○ |
| Factual reliability / source quality | ●●●○ | ●●●○ | ●●●○ | ●●●● | ●○○○ | ●●●○ | ●●●● |
| Creative quality (writing, ideas) | ●●●● | ●●●● | ●●●○ | ●●○○ | ●●○○ | ●●●○ | ●●●● |
| Prompt-quality tolerance (vague prompts) | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ | ●●●○ | ●●●○ |
| Safety / refusal balance | ●●●○ | ●○○○ | ●●●○ | ●●●○ | ●●○○ | ●●●○ | ●●●○ |
| Latest frontier model on plan | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Features & capabilities · 20% | 97 | 73 | 93 | 70 | 43 | 77 | 90 |
| Image generation | ✓ | ✗ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Image editing | ✓ | ✗ | ✓ | ~ | ~ | ~ | ✓ |
| Vision input | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Voice mode | ✓ | ~ | ✓ | ~ | ~ | ✓ | ~ |
| Video understanding | ~ | ✗ | ✓ | ✗ | ~ | ~ | ~ |
| Web search | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Answers with citations | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| File upload | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Data / spreadsheet analysis | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Code execution sandbox | ✓ | ✓ | ✓ | ~ | ✗ | ~ | ✓ |
| Custom assistants / GPTs | ✓ | ✓ | ✓ | ~ | ✗ | ~ | ✓ |
| Projects / Spaces | ✓ | ✓ | ~ | ✓ | ✗ | ~ | ✓ |
| Canvas / Artifacts | ✓ | ✓ | ✓ | ~ | ✗ | ~ | ✓ |
| Memory / personalization | ✓ | ✓ | ✓ | ~ | ~ | ✓ | ✓ |
| Scheduled / automated tasks | ✓ | ~ | ~ | ~ | ✗ | ~ | ~ |
| Workflow fit · 12% | 45 | 64 | 59 | 50 | 9 | 59 | 82 |
| Multiple providers in one interface | ✗ | ✗ | ✗ | ~ | ✗ | ~ | ✓ |
| Side-by-side answer comparison | ✗ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ |
| AI Collab (models debate / ensemble / co-operate) | ✗ | ✗ | ✗ | ✗ | ✗ | ✗ | ✓ |
| Document / file studio | ~ | ✓ | ~ | ~ | ✗ | ✓ | ✓ |
| Presentation creation (decks, PPTX) | ~ | ✓ | ✓ | ~ | ✗ | ✓ | ~ |
| Spreadsheet / Excel workflows | ✓ | ✓ | ✓ | ~ | ✗ | ✓ | ~ |
| OCR & large-file handling | ~ | ✓ | ✓ | ~ | ✗ | ~ | ✓ |
| Manual model switching / auto-router | ~ | ~ | ~ | ✓ | ✗ | ✗ | ✓ |
| Saved prompts / templates | ~ | ~ | ~ | ~ | ~ | ~ | ~ |
| Export (PDF, DOCX, MD, PPTX, CSV) | ~ | ✓ | ✓ | ~ | ✗ | ✓ | ~ |
| Team workspace | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Apps & ecosystem · 9% | 94 | 94 | 91 | 78 | 62 | 91 | 47 |
| iOS app | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Android app | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Desktop app (Win/Mac) | ✓ | ✓ | ~ | ✓ | ~ | ✓ | ~ |
| Browser extension | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ~ |
| Public API | ✓ | ✓ | ✓ | ✓ | ✓ | ~ | ✗ |
| API / SDK / docs quality | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●○○ | ●●●○ | ●○○○ |
| Office / Workspace integration | ~ | ~ | ✓ | ✗ | ✗ | ✓ | ✗ |
| 3rd-party integrations | ✓ | ✓ | ✓ | ~ | ~ | ✓ | ~ |
| Privacy & compliance · 8% | 81 | 81 | 75 | 44 | 19 | 94 | 88 |
| Training opt-out | ✓ | ✓ | ~ | ✓ | ~ | ✓ | ✓ |
| EU / regional data residency | ~ | ~ | ~ | ✗ | ✗ | ✓ | ✓ |
| Retention controls | ✓ | ✓ | ✓ | ~ | ✗ | ✓ | ✓ |
| Zero-data-retention option | ~ | ~ | ~ | ✗ | ✗ | ~ | ✓ |
| Certifications (SOC 2 / ISO 27001) | ✓ | ✓ | ✓ | ~ | ✗ | ✓ | ✓ |
| GDPR / DPA (AVV) | ✓ | ✓ | ✓ | ~ | ~ | ✓ | ✓ |
| HIPAA option | ~ | ~ | ~ | ✗ | ✗ | ✓ | ✗ |
| Broad regional availability | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Reliability & support · 7% | 89 | 84 | 93 | 86 | 59 | 89 | 77 |
| Response speed | ●●●○ | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ |
| First-token speed | ●●●○ | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ |
| Heavy-task speed | ●●●○ | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ |
| Uptime / stability | ●●●○ | ●●●○ | ●●●● | ●●●● | ●●○○ | ●●●● | ●●●○ |
| Low rate-limit frequency | ●●●○ | ●●●○ | ●●●● | ●●●● | ●●●○ | ●●●○ | ●●●○ |
| Answer consistency | ●●●● | ●●○○ | ●●●○ | ●●●● | ●●○○ | ●●●○ | ●●●○ |
| Help center | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Email / ticket support | ✓ | ✓ | ✓ | ~ | ~ | ✓ | ✓ |
| Enterprise support | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ~ |
| Public status page | ✓ | ✓ | ~ | ~ | ~ | ✓ | ~ |
| Data export / no lock-in | ✓ | ✓ | ✓ | ~ | ~ | ✓ | ✓ |
| Value & pricing · 7% | 68 | 78 | 88 | 86 | 53 | 78 | 90 |
| Entry price / month (USD) | $20 | $20 | $19.99 | $20 | $30 | $20 | $8.99 |
| Free tier | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ |
| Annual discount | ~ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Models included for the price | 1 | 1 | 1 | 4 | 1 | 1 | 5 |
| Student discount | ~ | ~ | ✓ | ~ | ✗ | ~ | ~ |
| Business & teams · 7% | 100 | 100 | 100 | 83 | 25 | 100 | 100 |
| Team plan | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| SSO / SCIM | ✓ | ✓ | ✓ | ✓ | ✗ | ✓ | ✓ |
| Admin console | ✓ | ✓ | ✓ | ~ | ✗ | ✓ | ✓ |
| Audit logs | ✓ | ✓ | ✓ | ~ | ✗ | ✓ | ✓ |
| Enterprise tier | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
| Shared workspaces | ✓ | ✓ | ✓ | ✓ | ~ | ✓ | ✓ |
Bold rows are category scores (0–100), computed from the feature rows beneath them. ✓ = yes, ~ = partial/limited, ✗ = no. Upload sizes and prices are approximate — verify officially.
Category by category
Who wins each category, and why
Each category scored from its parameters and weighted. Winner and full ranking in each.
Intelligence & model quality
Weight: 30%Reasoning, math, coding, research, agentic tool use, long context, multimodal, multilingual, instruction following, factual reliability, creative quality, prompt tolerance and safety.
🏆 Category winner: ChatGPT
ChatGPT OpenAI
95
Gemini Google
88
MultipleChat Multi-model
88
Claude Anthropic
82
Perplexity Perplexity
75
Copilot Microsoft
75
Grok xAI
61
Scored from: General reasoning, Math & logic, Coding, Deep research mode, Agentic tasks / tool use, Long-context handling, Multimodal understanding, Multilingual ability, Instruction following / steerability, Factual reliability / source quality, Creative quality, Prompt-quality tolerance, Safety / refusal balance, Latest frontier model on plan
Features & capabilities
Weight: 20%Concrete capabilities: images, voice, video, web, citations, files, data, code, custom assistants, projects, canvas, memory, automation.
🏆 Category winner: ChatGPT
ChatGPT OpenAI
97
Gemini Google
93
MultipleChat Multi-model
90
Copilot Microsoft
77
Claude Anthropic
73
Perplexity Perplexity
70
Grok xAI
43
Scored from: Image generation, Image editing, Vision input, Voice mode, Video understanding, Web search, Answers with citations, File upload, Data / spreadsheet analysis, Code execution sandbox, Custom assistants / GPTs, Projects / Spaces, Canvas / Artifacts, Memory / personalization, Scheduled / automated tasks
Workflow fit
Weight: 12%How the tool fits real work: multi-model use, side-by-side compare, collaboration, document/presentation/spreadsheet studios, OCR, exports, team space.
🏆 Category winner: MultipleChat
MultipleChat Multi-model
82
Claude Anthropic
64
Gemini Google
59
Copilot Microsoft
59
Perplexity Perplexity
50
ChatGPT OpenAI
45
Grok xAI
9
Scored from: Multiple providers in one interface, Side-by-side answer comparison, AI Collab, Document / file studio, Presentation creation, Spreadsheet / Excel workflows, OCR & large-file handling, Manual model switching / auto-router, Saved prompts / templates, Export, Team workspace
Apps & ecosystem
Weight: 9%Apps (iOS, Android, desktop), browser extension, public API and integrations.
🏆 Category winner: ChatGPT
ChatGPT OpenAI
94
Claude Anthropic
94
Gemini Google
91
Copilot Microsoft
91
Perplexity Perplexity
78
Grok xAI
62
MultipleChat Multi-model
47
Scored from: iOS app, Android app, Desktop app, Browser extension, Public API, API / SDK / docs quality, Office / Workspace integration, 3rd-party integrations
Privacy & compliance
Weight: 8%Training opt-out, regional data residency, retention, certifications, GDPR/DPA, HIPAA and availability.
🏆 Category winner: Copilot
Copilot Microsoft
94
MultipleChat Multi-model
88
ChatGPT OpenAI
81
Claude Anthropic
81
Gemini Google
75
Perplexity Perplexity
44
Grok xAI
19
Scored from: Training opt-out, EU / regional data residency, Retention controls, Zero-data-retention option, Certifications, GDPR / DPA, HIPAA option, Broad regional availability
Reliability & support
Weight: 7%Speed (response, first-token, heavy tasks), uptime, rate limits, consistency, support and data export.
🏆 Category winner: Gemini
Gemini Google
93
ChatGPT OpenAI
89
Copilot Microsoft
89
Perplexity Perplexity
86
Claude Anthropic
84
MultipleChat Multi-model
77
Grok xAI
59
Scored from: Response speed, First-token speed, Heavy-task speed, Uptime / stability, Low rate-limit frequency, Answer consistency, Help center, Email / ticket support, Enterprise support, Public status page, Data export / no lock-in
Value & pricing
Weight: 7%Price, free tier, annual discount, models included and student discount.
🏆 Category winner: MultipleChat
MultipleChat Multi-model
90
Gemini Google
88
Perplexity Perplexity
86
Claude Anthropic
78
Copilot Microsoft
78
ChatGPT OpenAI
68
Grok xAI
53
Scored from: Entry price / month, Free tier, Annual discount, Models included for the price, Student discount
Business & teams
Weight: 7%Team plans, SSO, admin, audit logs, enterprise tier and shared workspaces.
🏆 Category winner: ChatGPT
ChatGPT OpenAI
100
Claude Anthropic
100
Gemini Google
100
Copilot Microsoft
100
MultipleChat Multi-model
100
Perplexity Perplexity
83
Grok xAI
25
Scored from: Team plan, SSO / SCIM, Admin console, Audit logs, Enterprise tier, Shared workspaces