AI model comparison chart
40 current models from every major provider, side by side: price per million tokens, context window, max output, benchmark scores and release date. Sort any column, filter, or tick up to four models to compare and share.
Data synced 2026-11-12 00:00 UTC · Download CSV · Methodology
Best AI model for each job
Cheapest frontier model
DeepSeek · $0.99/M blended
Largest context window
OpenAI · 1.1M tokens
Best open-weight model
Moonshot AI · Newest open-weight release
Newest release
Google DeepMind · Released 2026-11-10
Computed from the table below, so these move when the data does.
Compare price, context and benchmarks
40 of 40 models. Tick up to 4 to compare side by side.
| Compare | Open | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Anthropic | Claude Fable 5.2 | 2026-11-09 | 1M | 128K | $8.00 | $40.00 | $8.00 | n/a | n/a | n/a | No | |
| Anthropic | Claude Haiku 5.5 | 2026-10-07 | 1M | 128K | $0.10 | $0.50 | $0.10 | n/a | n/a | n/a | No | |
| Anthropic | Claude Sonnet 5.5 | 2026-09-28 | 1M | 128K | $2.00 | $10.00 | $2.00 | n/a | n/a | n/a | No | |
| Anthropic | Claude Opus 5.5 | 2026-09-22 | 1M | 128K | $4.00 | $20.00 | $4.00 | n/a | n/a | n/a | No | |
| OpenAI | GPT-6.1 Terra Pro | 2026-10-26 | 1.1M | 128K | $1.00 | $5.00 | $1.05 | n/a | n/a | n/a | No | |
| OpenAI | GPT-6.1 Terra | 2026-10-26 | 1.1M | 128K | $1.00 | $5.00 | $1.05 | n/a | n/a | n/a | No | |
| OpenAI | GPT-6.1 Sol | 2026-09-29 | 1.1M | 128K | $2.00 | $10.00 | $2.10 | n/a | n/a | n/a | No | |
| OpenAI | GPT-6.1 Sol Pro | 2026-09-29 | 1.1M | 128K | $2.00 | $10.00 | $2.10 | n/a | n/a | n/a | No | |
| Google DeepMind | Gemini 4 Carbon | 2026-11-10 | 1M | 33K | $2.00 | $10.00 | $2.10 | n/a | n/a | n/a | No | |
| Google DeepMind | Gemini 3.8 Flash | 2026-09-02 | 1M | 66K | $0.75 | $3.75 | $0.79 | n/a | n/a | n/a | No | |
| Google DeepMind | Gemini 3.5 Flash Lite | 2026-07-21 | 1M | 66K | $0.30 | $2.50 | $0.31 | n/a | n/a | n/a | No | |
| Google DeepMind | Gemma 4 26B A4B IT | 2026-04-02 | 262K | 33K | $0.080 | $0.10 | $0.021 | n/a | n/a | n/a | Yes | |
| Meta AI | Muse Ember | 2026-11-09 | 1M | 944K | $0.10 | $0.50 | $0.10 | n/a | n/a | n/a | No | |
| Meta AI | Muse Ember | 2026-11-09 | 1M | 524K | $0.10 | $0.50 | $0.10 | n/a | n/a | n/a | No | |
| Meta AI | Muse Spark Open | 2026-10-15 | 1M | 262K | $1.25 | $4.25 | $1.31 | n/a | n/a | n/a | Yes | |
| Meta AI | Muse Wonder | 2026-10-12 | 1M | 944K | $4.00 | $20.00 | $4.19 | n/a | n/a | n/a | No | |
| xAI | Grok 4.7 | 2026-09-21 | 500K | 500K | $2.00 | $6.00 | $1.00 | n/a | n/a | n/a | No | |
| xAI | Grok Build 0.1 | 2026-04-16 | 256K | 256K | $1.00 | $2.00 | $0.26 | n/a | n/a | n/a | No | |
| DeepSeek | DeepSeek V4.1 Flash 1017 | 2026-10-17 | 1M | 393K | $0.15 | $0.60 | $0.16 | n/a | n/a | n/a | Yes | |
| DeepSeek | DeepSeek V4 Pro | 2026-08-13 | 1M | 393K | $0.66 | $1.98 | $0.69 | n/a | n/a | n/a | Yes | |
| DeepSeek | DeepSeek V4 Pro 0813 | 2026-08-12 | 1M | 944K | $0.11 | $4.44 | $0.12 | n/a | n/a | n/a | Yes | |
| DeepSeek | DeepSeek V3.2 | 2025-12-01 | 164K | 147K | $0.26 | $0.38 | $0.043 | n/a | n/a | n/a | Yes | |
| Alibaba Qwen | Qwen3.8 Max Prime | 2026-09-23 | 1M | 131K | $4.00 | $12.00 | $4.00 | n/a | n/a | n/a | No | |
| Alibaba Qwen | Qwen3.8 Omni Flash | 2026-09-17 | 1M | 131K | $0.11 | $0.38 | $0.11 | n/a | n/a | n/a | No | |
| Alibaba Qwen | Qwen3.8 Max (0902) | 2026-09-03 | 1M | 131K | $2.00 | $6.00 | $2.00 | n/a | n/a | n/a | No | |
| Alibaba Qwen | Qwen3.8 27B | 2026-08-14 | 1M | 236K | $0.15 | $1.88 | $0.16 | n/a | n/a | n/a | Yes | |
| Mistral AI | Mistral Large 4.0.8 | 2026-10-06 | 1M | 262K | $0.70 | $2.10 | $0.73 | n/a | n/a | n/a | No | |
| Mistral AI | Mistral Medium 3.5 | 2026-04-29 | 262K | 262K | $1.50 | $7.50 | $0.39 | n/a | n/a | n/a | Yes | |
| Mistral AI | Mistral Small 4 | 2026-03-16 | 262K | 256K | $0.15 | $0.60 | $0.039 | n/a | n/a | n/a | Yes | |
| Mistral AI | Devstral 2 | 2025-12-09 | 262K | 262K | $0.40 | $2.00 | $0.10 | n/a | n/a | n/a | Yes | |
| Moonshot AI | Kimi K3.1 Pro | 2026-10-23 | 1M | 1M | $2.00 | $10.00 | $2.10 | n/a | n/a | n/a | Yes | |
| Moonshot AI | Kimi K2.7 Code | 2026-06-12 | 262K | 262K | $0.95 | $4.00 | $0.25 | n/a | n/a | n/a | Yes | |
| Z.AI | GLM 5.3 FlashX Legal | 2026-11-02 | 1M | 131K | $0.37 | $1.25 | $0.39 | n/a | n/a | n/a | No | |
| Z.AI | GLM 5.3 Prime | 2026-09-23 | 1M | 131K | $2.80 | $8.80 | $2.80 | n/a | n/a | n/a | No | |
| Z.AI | GLM-5.3-FlashX | 2026-09-18 | 1M | 131K | $0.37 | $1.25 | $0.37 | n/a | n/a | n/a | Yes | |
| Z.AI | GLM-5.3 | 2026-08-14 | 1M | 131K | $1.40 | $4.40 | $1.40 | n/a | n/a | n/a | Yes | |
| Amazon | Nova 2 Lite | 2025-12-02 | 1M | 66K | $0.30 | $2.50 | $0.30 | n/a | n/a | n/a | No | |
| Amazon | Nova Premier 1.0 | 2025-10-31 | 1M | 32K | $2.50 | $12.50 | $2.50 | n/a | n/a | n/a | No | |
| Amazon | Nova Pro 1.0 | 2024-12-05 | 300K | 5K | $0.80 | $3.20 | $0.24 | n/a | n/a | n/a | No | |
| Amazon | Nova Micro 1.0 | 2024-12-05 | 128K | 5K | $0.035 | $0.14 | $0.004 | n/a | n/a | n/a | No |
* Self-reported by the provider. Unmarked scores come from public leaderboards.
Price vs SWE-bench score
Largest context windows
How this chart is built
- One row per current model in each active family, at most four per provider, ordered by provider.
- Prices are list prices in USD per 1M tokens. “Fill window” is the input cost of one full context window.
- Benchmark scores prefer public leaderboards; numbers a provider reports about its own model are marked with an asterisk.
- Advertised context is not effective context. Long-context recall usually degrades well before the advertised limit.
Sources and sync cadence are on the methodology page. For release dates and what ships next, see the AI model release tracker. Claude-specific changes live in the Claude Code changelog. Studying for the Claude certification? Try Claude Prep, or drill commands with /cards.
Prefer email?
New releases, straight to your inbox
One short email when a new AI model releases release ships. Skim what changed and get back to work.
No spam. Unsubscribe anytime. We only email when something new ships.
FAQ
AI model comparison FAQ
Which AI model is best for coding?+
Sort the table by the SWE-bench column. Benchmarks are a guide, not a verdict: try the top two or three on your own repository.
Which AI model has the largest context window?+
GPT-6.1 Terra Pro advertises 1.1M tokens. Effective context, where recall stays reliable, is usually well below the advertised number, so test long inputs before relying on them.
What is the cheapest GPT-class model?+
Among flagship-tier models, DeepSeek V4 Pro is currently cheapest at $0.99/M blended. Use the price filter to see every model under $1 or $3 per million input tokens.
Open-source vs closed AI models: which should I pick?+
Open-weight models can be self-hosted, fine-tuned and run without sending data to a provider, and they are often much cheaper through third-party hosts. Closed models still lead most frontier benchmarks. Tick "Open weights only" to compare the open field on its own.
How often is this chart updated?+
The catalog syncs from public model catalogs and leaderboards every 30 minutes, and this page refreshes within ten minutes of each sync. Prices are list prices in USD per 1M tokens.