# Free Reasoning & Chain-of-Thought LLM APIs

> 4 providers · 8 free models · Last updated: 01 Nov 2026

## Quick Answer

8 free Reasoning & Chain-of-Thought LLM APIs models across 4 providers including OpenCode Zen, OpenRouter, and Cloudflare Workers AI. Verified rate limits, context windows, and OpenAI-compatible endpoints. Last updated: 01 Nov 2026.

Free reasoning APIs offering extended thinking and step-by-step verification, featuring Nemotron Reasoning, Inkling, and DeepSeek R1 distills.

## Quick Connect

Chain-of-thought reasoning model with 1M context.

- **Base URL:** `https://openrouter.ai/api/v1`
- **Model:** `thinkingmachines/inkling:free`
- **Provider:** [OpenRouter](/providers/openrouter/)

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key="your-api-key",  # from OpenRouter
)

response = client.chat.completions.create(
    model="thinkingmachines/inkling:free",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
```

## Rate Limits & Model Inventory

| Model | Provider | Cost | Context | RPM | TPM | RPD | Day tokens | Verified |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Nemotron 3 Ultra Free | [OpenCode Zen](/providers/opencode-zen/) | $0 (promo) | 256K - 1M | Not published | Not published | Not published | Not published | 2026-11-01 |
| nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free | [OpenRouter](/providers/openrouter/) | $0 | 250K | 20 | Provider-dependent | 50 (below $10 credits) / 1,000 ($10+ credits) | Not published | 2026-11-01 |
| nvidia/nemotron-3-ultra-550b-a55b:free | [OpenRouter](/providers/openrouter/) | $0 | 1000000 | 20 | Provider-dependent | 50 (below $10 credits) / 1,000 ($10+ credits) | Not published | 2026-11-01 |
| thinkingmachines/inkling-small:free | [OpenRouter](/providers/openrouter/) | $0 | 1024K | 20 | Provider-dependent | 50 (below $10 credits) / 1,000 ($10+ credits) | Not published | 2026-11-01 |
| thinkingmachines/inkling:free | [OpenRouter](/providers/openrouter/) | $0 | 1024K | 20 | Provider-dependent | 50 (below $10 credits) / 1,000 ($10+ credits) | Not published | 2026-11-01 |
| ~80 free models (Llama 3.x/4, Qwen, Gemma, DeepSeek-R1 distills, FLUX, Whisper, BGE embeddings) | [Cloudflare Workers AI](/providers/cloudflare-workers-ai/) | $0 | 8K-128K+ (model-dependent) | 300 text gen (720-1,500 other tasks) | Model-dependent | Not published | 10,000 neurons/day | 2026-08-30 |
| @cf/deepseek-ai/deepseek-r1-distill-qwen-32b | [Cloudflare Workers AI](/providers/cloudflare-workers-ai/) | $0 | 128K | 300 | Model-dependent | Not published | 10,000 neurons/day | 2026-10-29 |
| deepseek-ai/DeepSeek-R1-Distill-Qwen-7B | [SiliconFlow](/providers/siliconflow/) | $0 | 32K | 1000 | 40K | 100 | Not published | 2026-10-29 |

Web page: https://freeinference.dev/free-reasoning-api/
Machine-readable JSON: /data/providers.json
Agent summary: /llms.txt

Numbers cross-checked against official docs and live probes as of 2026-11-01. Limits change without notice — this is a map, not a contract.