> ## Documentation Index > Fetch the complete documentation index at: https://docs.neosantara.xyz/llms.txt > Use this file to discover all available pages before exploring further. # Model Reasoning > Access internal chain-of-thought traces from reasoning models via reasoning_content. Reasoning models such as `deepseek-r1` and `deepseek-v4.1-flash` produce an internal chain-of-thought before emitting their final answers. Neosantara surfaces this thinking process in a dedicated `reasoning_content` field on `/v1/chat/completions`. ```python Python (Streaming / OpenAI SDK) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}} from openai import OpenAI import os client = OpenAI( base_url="https://api.neosantara.xyz/v1", api_key=os.environ["NEOSANTARA_API_KEY"] ) stream = client.chat.completions.create( model="deepseek-r1", messages=[ {"role": "user", "content": "How many 'r' letters are in the word 'strawberry'?"} ], stream=True ) print("=== REASONING PROCESS ===") for chunk in stream: delta = chunk.choices[0].delta # Extract reasoning tokens reasoning = getattr(delta, "reasoning_content", None) if reasoning: print(reasoning, end="", flush=True) # Extract final public content content = delta.content or "" if content: print(content, end="", flush=True) print() ``` ```python Python (Non-Streaming) icon="python" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}} from openai import OpenAI import os client = OpenAI( base_url="https://api.neosantara.xyz/v1", api_key=os.environ["NEOSANTARA_API_KEY"] ) response = client.chat.completions.create( model="deepseek-r1", messages=[ {"role": "user", "content": "How many 'r' letters are in the word 'strawberry'?"} ] ) message = response.choices[0].message # Internal thinking trace if hasattr(message, "reasoning_content") and message.reasoning_content: print(f"[Thinking]:\n{message.reasoning_content}\n") # Final public answer print(f"[Answer]:\n{message.content}") ``` ```bash cURL icon="terminal" theme={"theme":{"light":"ayu-dark","dark":"catppuccin-latte"}} curl -X POST https://api.neosantara.xyz/v1/chat/completions \ -H "Authorization: Bearer $NEOSANTARA_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "deepseek-r1", "messages": [ {"role": "user", "content": "How many r in strawberry?"} ] }' ``` ## Billing Rules for Reasoning Tokens Neosantara adheres to strict billing integrity principles: * **Delivered Output Is Billable:** Reasoning tokens delivered to the client (via `reasoning_content`) are counted as output tokens in `usage.completion_tokens`. * **Usage Breakdown:** The usage payload includes `completion_tokens_details.reasoning_tokens` allowing exact analysis of thinking token volume. * **Model Integrity:** Billing is calculated at the exact public rate of the model requested. ## Available Reasoning Models | Model ID | Provider | Context Window | Capabilities | Pricing (Input/Output per 1M) | | :---------------------------------- | :---------- | :------------- | :----------------------------- | :---------------------------- | | `agnes-2.5-flash` | Agnes AI | 1M tokens | Tools, Vision, Reasoning, JSON | Free / Included | | `claude-4-5-haiku` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $1 / $5 | | `claude-4.5-opus` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $5 / $25 | | `claude-4.5-sonnet` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $3 / $15 | | `claude-fable-5` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $10 / $50 | | `claude-opus-4-6` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $5 / $25 | | `claude-opus-4-7` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $5 / $25 | | `claude-opus-4-8` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $5 / $25 | | `claude-opus-5` | Anthropic | 1M tokens | Tools, Vision, Reasoning | $5 / $25 | | `claude-sonnet-4-6` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $3 / $15 | | `claude-sonnet-5` | Anthropic | 200k tokens | Tools, Vision, Reasoning, JSON | $3 / $15 | | `deepseek-r1` | DeepSeek | 128k tokens | Reasoning, JSON | $1.35 / $5.4 | | `deepseek-v4-flash` | DeepSeek | 1M tokens | Tools, Reasoning | $0.22 / $0.66 | | `deepseek-v4-flash-0731` | DeepSeek | 1M tokens | Tools, Reasoning, JSON | $0.14 / $0.28 | | `deepseek-v4-pro` | DeepSeek | 1M tokens | Tools, Reasoning | $0.66 / $1.98 | | `deepseek-v4-pro-0813` | DeepSeek | 1M tokens | Tools, Reasoning, JSON | $1.32 / $3.96 | | `deepseek-v4.1-flash` | DeepSeek | 1M tokens | Tools, Vision, Reasoning, JSON | $0.15 / $0.6 | | `gemini-3.1-flash-lite` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $0.25 / $1.5 | | `gemini-3.1-pro` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $2 / $12 | | `gemini-3.5-flash` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $1.5 / $9 | | `gemini-3.6-flash` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $1.5 / $7.5 | | `gemini-3.7-flash` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $1.5 / $7.5 | | `gemini-3.8-flash` | Google | 1M tokens | Tools, Vision, Reasoning, JSON | $0.75 / $3.75 | | `ling-3.0-flash-fin` | InclusionAI | 256k tokens | Tools, Reasoning | Free / Included | | `longcat-2.0` | Meituan | 1M tokens | Tools, Reasoning, JSON | Free / Included | | `muse-glimmer-30b` | Meta | 131k tokens | Tools, Vision, Reasoning, JSON | $0.35 / $1.5 | | `muse-spark-1.1` | Meta | 1M tokens | Tools, Vision, Reasoning | $1.25 / $4.25 | | `minimax-m2.7` | MiniMax | 204k tokens | Tools, Reasoning, JSON | $0.3 / $1.2 | | `minimax-m2.7-free` | MiniMax | 196k tokens | Tools, Reasoning, JSON | Free / Included | | `minimax-m3` | MiniMax | 1M tokens | Tools, Vision, Reasoning, JSON | $0.6 / $1.2 | | `minimax-m3-free` | MiniMax | 1M tokens | Tools, Vision, Reasoning, JSON | Free / Included | | `kimi-k2-thinking` | Moonshot | 256k tokens | Tools, Reasoning | $0.6 / $2.5 | | `kimi-k2.5` | Moonshot | 256k tokens | Tools, Vision, Reasoning | $0.6 / $3 | | `kimi-k2.6` | Moonshot | 262k tokens | Tools, Vision, Reasoning, JSON | $0.95 / $4 | | `kimi-k3` | Moonshot | 1M tokens | Tools, Vision, Reasoning, JSON | $3 / $15 | | `garda-core` | Neosantara | 1M tokens | Tools, Reasoning, JSON | Rp 300 / Rp 1,500 | | `llama-3.3-nemotron-super-49b-v1.5` | NVIDIA | 132k tokens | Tools, Reasoning, JSON | Rp 4,560 / Rp 31,991 | | `gpt-5-nano` | OpenAI | 128k tokens | Tools, Vision, Reasoning, JSON | Rp 810 / Rp 6,485 | | `gpt-5.4` | OpenAI | 270k tokens | Tools, Vision, Reasoning, JSON | $2.5 / $15 | | `gpt-5.4-mini` | OpenAI | 128k tokens | Tools, Vision, Reasoning, JSON | $0.75 / $4.5 | | `gpt-5.4-nano` | OpenAI | 400k tokens | Tools, Vision, Reasoning, JSON | Rp 3,240 / Rp 20,250 | | `gpt-5.5` | OpenAI | 270k tokens | Tools, Vision, Reasoning, JSON | $5 / $30 | | `gpt-5.6-luna` | OpenAI | 272k tokens | Tools, Reasoning | $0.2 / $1.2 | | `gpt-5.6-sol` | OpenAI | 272k tokens | Tools, Reasoning | $5 / $30 | | `gpt-5.6-terra` | OpenAI | 272k tokens | Tools, Vision, Reasoning | $2 / $12 | | `gpt-oss-120b` | OpenAI | 131k tokens | Tools, Reasoning, JSON | $0.15 / $0.6 | | `gpt-oss-20b` | OpenAI | 131k tokens | Tools, Reasoning, JSON | Rp 400 / Rp 1,600 | | `laguna-s-2.1` | Poolside | 262k tokens | Tools, Reasoning | Free / Included | | `laguna-xs-2.1` | Poolside | 262k tokens | Tools, Reasoning | Free / Included | | `bonsai-27b` | Prism ML | 262k tokens | Tools, Vision, Reasoning, JSON | Free / Included | | `step-3.5-flash` | StepFun | 256k tokens | Tools, Reasoning, JSON | Rp 3,000 / Rp 9,000 | | `grok-4.1-fast-reasoning` | xAI | 1M tokens | Tools, Reasoning | $1.25 / $2.5 | | `grok-4.5` | xAI | 500k tokens | Tools, Vision, Reasoning, JSON | $2 / $6 | | `mimo-v2.5` | Xiaomi | 1M tokens | Tools, Vision, Reasoning | $0.14 / $0.28 | | `mimo-v2.5-pro` | Xiaomi | 1M tokens | Tools, Vision, Reasoning, JSON | $0.435 / $0.87 | | `glm-4.5-flash` | zAI | 128k tokens | Tools, Reasoning | Rp 250 / Rp 750 | | `glm-4.6` | zAI | 128k tokens | Vision, Reasoning | Rp 10,000 / Rp 30,000 | | `glm-4.6-plus` | zAI | 128k tokens | Vision, Reasoning | Rp 10,000 / Rp 30,000 | | `glm-4.6v-flash` | zAI | 128k tokens | Tools, Vision, Reasoning | Rp 250 / Rp 750 | | `glm-4.7` | zAI | 200k tokens | Tools, Reasoning | Rp 10,000 / Rp 30,000 | | `glm-4.7-flash` | zAI | 1M tokens | Tools, Reasoning | Free / Included | | `glm-5.3` | zAI | 1M tokens | Tools, Vision, Reasoning, JSON | $1.4 / $4.4 | | `glm-5.3-flash` | zAI | 1M tokens | Tools, Vision, Reasoning, JSON | $0.15 / $0.5 | ## Next Steps | Task | Guide | | :-------------------------- | :------------------------------------------------------------ | | Anthropic Extended Thinking | [Extended Thinking](/en/gateway/anthropic-messages/thinking) | | Responses API Reasoning | [Responses Reasoning](/en/gateway/responses-api/reasoning) | | Streaming Responses | [Streaming Responses](/en/gateway/chat-completions/streaming) | ## Related topics - [Chat Completions](/en/gateway/chat-completions.md) - [Anthropic Extended Thinking](/en/gateway/anthropic-messages/thinking.md) - [Responses API Reasoning](/en/gateway/responses-api/reasoning.md) - [Model Catalog](/en/gateway/models.md)