AI Model Comparison
Compare 41 frontier AI models on price, context and independent benchmarks.
Prices updated . Benchmarks updated .
41 models
| Compare | Model | You send / 1M | It replies / 1M | Context | Good to know | Launched | ECI | GPQA | ARC-AGI-2 | HLE | AIME | SimpleQA | CritPt | Arena | Arena web | Tokens/s | Latency | It replies / 1M | Context |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GPT-6 AstraOpenAI | $10.00 | $50.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-04 | 169.1 | 95.8% | 95.0% | 54.8% | 100.0% | 75.6% | 31.7% | 1,441 | 1,773 | n/a | n/a | $50.00 | 1.05M | |
| Claude Opus 5.5Anthropic | $4.00 | $20.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-09-22 | 168.4 | 90.6% | 93.3% | n/a | 100.0% | 72.2% | 31.7% | 1,516 | 1,810 | n/a | n/a | $20.00 | 1M | |
| GPT-6.1 TerraOpenAI | $1.00 | $5.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-10-26 | 168.1 | 94.3% | n/a | n/a | 100.0% | 57.3% | 28.0% | n/a | 1,673 | n/a | n/a | $5.00 | 1.05M | |
| GPT-6.1 SolOpenAI | $2.00 | $10.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-29 | 167.8 | 95.4% | 94.2% | n/a | 100.0% | 73.9% | 31.7% | 1,446 | 1,751 | n/a | n/a | $10.00 | 1.05M | |
| Claude Sonnet 5.5Anthropic | $2.00 | $10.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-09-28 | 166.7 | 95.6% | 83.9% | 46.1% | 100.0% | 46.5% | 31.4% | 1,473 | 1,770 | n/a | n/a | $10.00 | 1M | |
| Claude Fable 5.1Anthropic | $10.00 | $50.00 | 1M | No training · Images · Reasoning · Tools | 2026-09-01 | 166.4 | n/a | 90.0% | 46.5% | 100.0% | 70.8% | 31.1% | 1,514 | 1,746 | n/a | n/a | $50.00 | 1M | |
| Gemini 4 ArgonGoogle | $2.00 | $10.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-10-13 | 166.1 | 96.4% | 91.3% | n/a | 100.0% | 76.0% | 27.1% | 1,529 | 1,679 | n/a | n/a | $10.00 | 1.05M | |
| Claude Opus 5Anthropic | $5.00 | $25.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-07-24 | 164.5 | 93.9% | 90.4% | n/a | 98.9% | 59.9% | 29.1% | 1,508 | 1,691 | n/a | n/a | $25.00 | 1M | |
| GPT-6 SolOpenAI | $2.00 | $10.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-22 | 163.8 | 94.3% | 89.6% | n/a | 100.0% | 60.7% | 30.9% | 1,397 | 1,678 | n/a | n/a | $10.00 | 1.05M | |
| Claude Fable 5Anthropic | $10.00 | $50.00 | 1M | No training · Images · Reasoning · Tools | 2026-06-09 | 163.0 | 85.9% | 89.2% | n/a | 100.0% | 70.7% | 28.6% | 1,492 | 1,624 | n/a | n/a | $50.00 | 1M | |
| Muse WonderMeta | $4.00 | $20.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-10-12 | 161.7 | n/a | n/a | n/a | n/a | n/a | 28.3% | 1,499 | 1,652 | n/a | n/a | $20.00 | 1.05M | |
| Claude Haiku 5.5Anthropic | $0.100 | $0.500 | 1M | ZDR · Images · Reasoning · Tools | 2026-10-07 | 160.4 | 89.7% | 61.3% | 30.1% | 98.9% | 23.8% | 22.6% | 1,401 | 1,589 | n/a | n/a | $0.500 | 1M | |
| Claude Opus 4.8Anthropic | $5.00 | $25.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-05-28 | 159.6 | 91.0% | 72.1% | n/a | 98.3% | 53.0% | 20.9% | 1,461 | 1,556 | n/a | n/a | $25.00 | 1M | |
| GPT-5.5OpenAI | $5.00 | $30.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-04-23 | 159.4 | 90.7% | 85.0% | n/a | 84.4% | 63.0% | 27.1% | 1,472 | n/a | n/a | n/a | $30.00 | 1M | |
| DeepSeek V4.1 Flash 1017DeepSeek · Open weights | $0.300 | $1.20 | 1M | ZDR · Images · Reasoning · Tools | 2026-10-17 | 158.8 | 93.2% | 78.1% | n/a | 99.4% | 42.5% | 24.0% | n/a | n/a | n/a | n/a | $1.20 | 1M | |
| Kimi K3Moonshot · Open weights | $0.192 | $16.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-07-16 | 158.3 | 93.1% | 60.4% | n/a | 97.2% | 50.6% | 23.4% | 1,476 | 1,656 | n/a | n/a | $16.00 | 1.05M | |
| Gemini 3.7 FlashGoogle | $0.750 | $3.75 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-08-13 | 158.2 | 94.8% | 84.6% | n/a | 97.2% | 69.2% | 14.3% | 1,489 | 1,590 | n/a | n/a | $3.75 | 1.05M | |
| Gemini 3.8 FlashGoogle | $0.825 | $4.13 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-02 | 157.6 | 95.4% | 89.2% | 44.5% | 98.9% | 69.7% | 18.3% | 1,498 | 1,583 | n/a | n/a | $4.13 | 1.05M | |
| Muse Spark 1.3Meta | $1.25 | $4.25 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-02 | 157.4 | n/a | n/a | n/a | 99.2% | n/a | 26.0% | 1,491 | 1,656 | n/a | n/a | $4.25 | 1.05M | |
| Claude Sonnet 5Anthropic | $2.00 | $10.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-06-30 | 157.0 | 90.5% | n/a | n/a | 94.7% | 33.7% | 16.9% | 1,443 | 1,540 | n/a | n/a | $10.00 | 1M | |
| Grok 4.6xAI | $2.00 | $6.00 | 500K | ZDR · Images · Reasoning · Tools | 2026-08-12 | 156.9 | 94.0% | 67.1% | n/a | 99.2% | 49.3% | 19.7% | 1,429 | 1,616 | n/a | n/a | $6.00 | 500K | |
| GPT-6 LunaOpenAI | $0.100 | $0.500 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-09-22 | 156.8 | 90.5% | 59.3% | n/a | 98.9% | 41.4% | 19.4% | 1,389 | 1,579 | n/a | n/a | $0.500 | 1.05M | |
| Qwen3.8 Max (0902)Qwen | $2.00 | $6.00 | 984K | No training · Images · Reasoning · Tools | 2026-09-02 | 156.2 | 92.3% | n/a | n/a | 100.0% | 47.3% | 17.7% | n/a | 1,676 | n/a | n/a | $6.00 | 984K | |
| GLM 5.3Z.ai · Open weights | $0.070 | $1.40 | 1M | ZDR · Reasoning · Tools | 2026-08-14 | 156.2 | 90.9% | n/a | n/a | 91.1% | 41.0% | 19.1% | 1,474 | 1,623 | n/a | n/a | $1.40 | 1M | |
| DeepSeek V4.1 FlashDeepSeek · Open weights | $0.140 | $0.420 | 1M | ZDR · Images · Reasoning · Tools | 2026-09-10 | 156.1 | 89.8% | 72.9% | n/a | 98.3% | 42.9% | 14.3% | n/a | n/a | n/a | n/a | $0.420 | 1M | |
| DeepSeek V4 Pro 0813DeepSeek · Open weights | $0.111 | $4.44 | 1M | ZDR · Reasoning · Tools | 2026-08-13 | 155.9 | 91.7% | 61.3% | n/a | 98.6% | 52.9% | 18.0% | n/a | n/a | n/a | n/a | $4.44 | 1M | |
| Gemini 3.1 Pro PreviewGoogle | $2.00 | $12.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-02-19 | 155.2 | 94.4% | 77.1% | 46.4% | 95.6% | 73.5% | 17.7% | 1,480 | 1,447 | n/a | n/a | $12.00 | 1.05M | |
| Muse Spark OpenMeta · Open weights | $1.25 | $4.25 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-10-15 | 155.1 | n/a | 57.5% | n/a | n/a | n/a | 14.9% | n/a | 1,526 | n/a | n/a | $4.25 | 1.05M | |
| Gemini 3.5 FlashGoogle | $1.50 | $9.00 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-05-19 | 155.0 | 92.8% | 72.1% | n/a | 95.6% | 66.2% | 13.1% | 1,482 | 1,499 | n/a | n/a | $9.00 | 1.05M | |
| Grok 4.7xAI | $2.00 | $6.00 | 500K | ZDR · Images · Reasoning · Tools | 2026-09-21 | 154.6 | 92.7% | 61.4% | n/a | 98.1% | 56.0% | 18.0% | 1,398 | 1,637 | n/a | n/a | $6.00 | 500K | |
| Grok 4.5xAI | $2.00 | $6.00 | 500K | ZDR · Images · Reasoning · Tools | 2026-07-08 | 154.4 | 93.4% | 52.6% | n/a | 97.8% | 48.3% | 15.4% | 1,448 | 1,553 | n/a | n/a | $6.00 | 500K | |
| Qwen3.7 MaxQwen | $1.48 | $4.42 | 1M | No training · Reasoning · Tools | 2026-05-21 | 154.2 | 90.9% | n/a | n/a | 95.6% | 55.8% | 13.4% | n/a | n/a | n/a | n/a | $4.42 | 1M | |
| GLM 5.3 FlashZ.ai · Open weights | $0.040 | $0.078 | 1M | ZDR · Images · Reasoning · Tools | 2026-08-26 | 153.2 | 90.2% | 65.8% | n/a | 93.9% | n/a | 15.4% | 1,473 | 1,607 | n/a | n/a | $0.078 | 1M | |
| Qwen3.8 27BQwen · Open weights | $0.150 | $1.88 | 262K | ZDR · Images · Reasoning · Tools | 2026-08-14 | 151.8 | n/a | 42.4% | n/a | n/a | n/a | 5.4% | 1,438 | 1,590 | n/a | n/a | $1.88 | 262K | |
| Mistral Large 4Mistral | $0.700 | $2.10 | 1M | ZDR · Images · Reasoning · Tools | 2026-10-06 | 150.6 | 88.1% | n/a | n/a | 98.3% | 20.0% | 10.6% | 1,430 | 1,544 | n/a | n/a | $2.10 | 1M | |
| MiniMax M3MiniMax · Open weights | $0.300 | $1.20 | 1M | ZDR · Images · Reasoning · Tools | 2026-06-01 | 146.9 | 90.9% | n/a | n/a | 71.1% | n/a | 3.7% | 1,431 | 1,483 | n/a | n/a | $1.20 | 1M | |
| Gemini 4 CarbonPreviewGoogle | $2.00 | $10.00 | 1M | ZDR · Images · Reasoning · Tools | 2026-10-28 | n/a | n/a | n/a | n/a | n/a | n/a | 36.0% | 1,540 | 1,768 | n/a | n/a | $10.00 | 1M | |
| Particle-1Particle | $0.060 | $0.500 | 1.05M | Free · ZDR · Images · Reasoning · Tools | 2026-10-28 | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | $0.500 | 1.05M | |
| Kimi K3.1 ProMoonshot · Open weights | $0.240 | $8.40 | 1.05M | ZDR · Images · Reasoning · Tools | 2026-10-23 | n/a | n/a | n/a | n/a | n/a | n/a | 24.6% | n/a | 1,695 | n/a | n/a | $8.40 | 1.05M | |
| MiniMax M3.1 FlashMiniMax · Open weights | $0.150 | $0.600 | 1M | ZDR · Images · Reasoning · Tools | 2026-10-19 | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | 1,582 | n/a | n/a | $0.600 | 1M | |
| Qwen3.8 FlashQwen | $0.150 | $0.470 | 1M | No training · Images · Reasoning · Tools | 2026-08-26 | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a | $0.470 | 1M |
Independent scores. ECI is the Epoch Capabilities Index. GPQA is GPQA Diamond. AIME is OTIS Mock AIME. Arena is LMArena Elo. Empty cells mean the current open dataset has no score for that model. Column names link to the benchmark guide.
Tokens per second and latency are via OpenRouter, last 30 minutes, and only when OpenRouter publishes a reading. Many models have none.
What is the best model right now?
GPT-6 Astra leads the frontier models on this page with an Epoch Capabilities Index of 169.1. Claude Opus 5.5 is next at 168.4. GPT-6 Astra costs $50.00 per million output tokens and reads 1.05M of context. Claude Opus 5.5 costs $20.00 per million output tokens and reads 1M of context.
What is the cheapest frontier model?
GLM 5.3 Flash has the lowest published output price among frontier models, at $0.078 per million tokens. Input tokens cost $0.040 per million and the context window is 1M. DeepSeek V4.1 Flash is the next lowest at $0.420 per million output tokens. A lower price is not a quality ranking. Compare the Epoch Capabilities Index before you choose.
Which open-weight models lead?
DeepSeek V4.1 Flash 1017 leads open-weight models that have a published Epoch Capabilities Index, at 158.8. It is from DeepSeek, costs $1.20 per million output tokens, and reads 1M. Kimi K3 from Moonshot follows at 158.3. Open weights means the lab published the weights. The license on the model page says what you can do with them.
FAQ
What is the best model right now?
GPT-6 Astra leads the frontier models on this page with an Epoch Capabilities Index of 169.1. Claude Opus 5.5 is next at 168.4. GPT-6 Astra costs $50.00 per million output tokens and reads 1.05M of context. Claude Opus 5.5 costs $20.00 per million output tokens and reads 1M of context.
What is the cheapest frontier model?
GLM 5.3 Flash has the lowest published output price among frontier models, at $0.078 per million tokens. Input tokens cost $0.040 per million and the context window is 1M. DeepSeek V4.1 Flash is the next lowest at $0.420 per million output tokens. A lower price is not a quality ranking. Compare the Epoch Capabilities Index before you choose.
Which open-weight models lead?
DeepSeek V4.1 Flash 1017 leads open-weight models that have a published Epoch Capabilities Index, at 158.8. It is from DeepSeek, costs $1.20 per million output tokens, and reads 1M. Kimi K3 from Moonshot follows at 158.3. Open weights means the lab published the weights. The license on the model page says what you can do with them.
Where do the benchmark scores come from?
Independent scores, including the Epoch Capabilities Index, come from Epoch AI. Arena Elo comes from the LMArena leaderboard dataset. Both are used under CC BY 4.0. The benchmark file date is 2026-10-31.
How fresh are the prices?
Token prices were fetched Nov 16, 2026, 2:31 AM ET. They are US dollars per million tokens at the current public rate and refresh about hourly. This page does not chart older prices.
How to read this table
Input price is what you pay for the tokens you send. Output price is what you pay for tokens the model writes, and it is usually several times higher. Context window is the most text a model can read in one request. The launch date is the lab's date when we have one, not the day a reseller listed the model.
The default Frontier view is a curated set. All adds the rest of the public catalogue from major labs. Open-weight and Under $1 narrow it further. Tick two to four rows to compare them side by side.
Want a monthly estimate? Use the AI API cost calculator.
Sources and licenses
- OpenRouter: Prices, context windows, modalities and the model list come from the OpenRouter public API. Data: OpenRouter.
- Vercel AI Gateway: The no-training flag comes from the Vercel AI Gateway model list.
- models.dev: Lab launch dates, open-weights flags and knowledge cutoffs come from models.dev. Copyright (c) 2025 models.dev, MIT License.
- Epoch AI: Independent benchmark scores and the Epoch Capabilities Index (ECI) come from Epoch AI's Benchmarking Hub, used under CC BY 4.0. Epoch AI, Capabilities & benchmarking, epoch.ai/benchmarks. Scores Epoch marks as external keep their original license, and the source is named on each score.
- LMArena: Arena Elo (text, webdev and vision) comes from the LMArena leaderboard dataset (lmarena-ai/leaderboard-dataset), CC BY 4.0, leaderboard date 2026-10-28.
- Lab-reported scores, when an editor has added them, are the lab's own numbers. Each one links to the source and is labeled lab-reported. They are not used to rank the lists.
- What a score measures: the benchmark guide.
