Language
Large language models
Chat, reasoning, coding, agents, and text generation.
What it is
A large language model (LLM) reads text (and often images, audio or files) and generates text: answers, code, summaries, structured JSON, or tool calls that let it act as an agent. It is the most general-purpose AI API and the backbone of chatbots, copilots and automation.
How pricing works
Billed per token (≈ ¾ of an English word). Input tokens (what you send) and output tokens (what the model writes) are priced separately, and output is usually 3–5× more expensive. We rank on output price; the detail page shows both, plus discounts such as prompt caching and batch processing (often 50% off).
We compare everything in one unit: USD per 1M output tokens.
Typical uses
- Chatbots and support assistants
- Coding assistants and code review
- Agents that call tools and APIs
- Summarisation, extraction and classification
- Drafting and rewriting content
What to compare
- Quality on reasoning/coding benchmarks
- Context window (how much text fits in one request)
- Output speed (tokens/second) and time-to-first-token
- Tool-use and structured-output reliability
Watch out for
- Cheapest per token is not cheapest per task: a weaker model may need more retries or longer prompts.
- Reasoning/'thinking' models bill hidden thinking tokens as output.
- Long-context requests can have a higher price tier.
All 25 APIs, cheapest first
Free to browse| API | $ / 1M out tokens | Context window | Max output | Input types | Quality | Price data | Compare |
|---|---|---|---|---|---|---|---|
| Amazon Nova Lite Amazon Web Services | $0.24 in $0.06 | 300K tokens | — | text, image, video | – | Unverified sample | |
| Qwen3.8 Flash Alibaba Cloud (Qwen) | $0.47 in $0.15 | — | — | — | – | Verified Oct 6, 2026 | |
| GPT-6 Luna OpenAI | $0.50 in $0.10 | 1,050K tokens | 128K tokens | text, image | 78 | Verified Oct 4, 2026 · aging | |
| GLM-5.3 Flash Z.ai (Zhipu GLM) | $0.50 in $0.15 | — | — | — | – | Verified Oct 6, 2026 | |
| Llama API: Llama 4 Maverick Meta (Llama API) | $0.85 in $0.27 | 1,000K tokens | — | text, image | – | Unverified sample | |
| DeepSeek V4.1 Flash DeepSeek | $1.20 in $0.30 | 1,000K tokens | 384K tokens | text, image | 82 | Verified Oct 4, 2026 | |
| Grok 4.3 xAI (Grok) | $2.50 in $1.25 | 1,000K tokens | — | text, image | – | Verified Oct 6, 2026 | |
| Amazon Nova Pro Amazon Web Services | $3.20 in $0.80 | 300K tokens | — | text, image, video | – | Unverified sample | |
| Gemini 3.8 Flash Google (Gemini / Vertex) | $3.75 in $0.75 | 1,000K tokens | 64K tokens | text, image, audio, video, pdf | 80 | Verified Oct 4, 2026 · aging | |
| Kimi K2.6 Moonshot AI (Kimi) | $4.00 in $0.95 | 262.144K tokens | — | — | – | Verified Oct 6, 2026 | |
| Azure OpenAI: GPT-6 Luna Microsoft Azure AI | $4.00 in $0.50 | — | — | — | – | Unverified sample | |
| Claude Haiku 4.5 Anthropic | $5.00 in $1.00 | 200K tokens | 64K tokens | text, image, pdf | 80 | Verified Oct 4, 2026 · aging | |
| Grok 4.7 xAI (Grok) | $6.00 in $2.00 | 500K tokens | — | text, image | – | Verified Oct 6, 2026 | |
| Qwen3.8 Max Alibaba Cloud (Qwen) | $6.00 in $2.00 | — | — | — | – | Verified Oct 6, 2026 | |
| Mistral Medium 3.5 Mistral AI | $7.50 in $1.50 | 262.144K tokens | 32K tokens | text, image | 76 | Verified Oct 4, 2026 | |
| Jamba Large AI21 Labs | $8.00 in $2.00 | 256K tokens | — | — | – | Unverified sample | |
| Claude Sonnet 5.5 Anthropic | $10.00 in $2.00 | 1,000K tokens | 128K tokens | text, image, pdf | 91 | Verified Oct 4, 2026 · aging | |
| GPT-6.1 Sol OpenAI | $10.00 in $2.00 | 1,050K tokens | 128K tokens | text, image | 91 | Verified Oct 4, 2026 · aging | |
| Command A Cohere | $10.00 in $2.50 | 256K tokens | — | — | – | Unverified sample | |
| Gemini 3.1 Pro Preview Google (Gemini / Vertex) | $12.00 in $2.00 | 1,000K tokens | 64K tokens | text, image, audio, video, pdf | 92 | Verified Oct 4, 2026 · aging | |
| Kimi K3 Moonshot AI (Kimi) | $15.00 in $3.00 | 1,048.576K tokens | — | — | – | Verified Oct 6, 2026 | |
| Sonar Pro Perplexity | $15.00 in $3.00 | — | — | text | – | Verified Oct 6, 2026 | |
| Claude Opus 5.5 Anthropic | $20.00 in $4.00 | 1,000K tokens | 128K tokens | text, image, pdf | 95 | Verified Oct 4, 2026 · aging | |
| Claude Fable 5.1 Anthropic | $50.00 in $10.00 | 1,000K tokens | 128K tokens | text, image, pdf | 98 | Verified Oct 4, 2026 · aging | |
| GPT-6 Astra OpenAI | $50.00 in $10.00 | 1,050K tokens | 128K tokens | text, image | 93 | Verified Oct 4, 2026 · aging |