321 models on CoreAI support native reasoning — they "think" through a problem step-by-step before answering, which dramatically improves accuracy on math, logic puzzles, competitive coding, and multi-step analysis. Reasoning-capable models are available from Meta, NVIDIA, Liquid, Google, Stepfun and more, listed newest-first below.
Reasoning mode trades speed for quality: responses take longer and consume more tokens, but for hard problems the difference is decisive. In CoreAI, models that support thinking show a "Think" chip in chat — tap it to enable deliberate reasoning for that message.
| Model | Provider | Context | Input $/1M | Output $/1M | Capabilities |
|---|---|---|---|---|---|
| Meta: Muse Spark Open | Meta | 1049K | $1.25 | $4.25 | Reasoning, Vision |
| NVIDIA: Nemotron 3.5 Super VL | NVIDIA | 262K | $0.15 | $0.50 | Reasoning, Vision |
| NVIDIA: Nemotron 3.5 Super VL (free) | NVIDIA | 1000K | $0 | $0 | Reasoning, Vision |
| LiquidAI: LFM2.5-24B-A2B | Liquid | 128K | $0.05 | $0.70 | Reasoning |
| Google: Gemini 4 Argon | 1049K | $2 | $10 | Reasoning, Vision | |
| Google: Gemini 4 Argon (batch) | 1049K | $1 | $5 | Reasoning, Vision | |
| Meta: Muse Wonder | Meta | 1049K | $4 | $20 | Reasoning, Vision |
| StepFun: Step 5 Preview | Stepfun | 1000K | $1 | $2.70 | Reasoning, Vision |
| Anthropic: Claude Haiku 5.5 | Anthropic | 1000K | $0.10 | $0.50 | Reasoning, Vision |
| Anthropic: Claude Haiku 5.5 (batch) | Anthropic | 1000K | $0.05 | $0.25 | Reasoning, Vision |
| Mistral: Mistral Large 4 | Mistral AI | 1049K | $0.68 | $2.09 | Reasoning, Vision |
| inclusionAI: Ling 3.1 Flash | Inclusionai | 262K | $0.07 | $0.22 | Reasoning |
| OpenAI: GPT-6.1 Sol Pro | OpenAI | 1050K | $2 | $10 | Reasoning, Vision |
| OpenAI: GPT-6.1 Sol Pro (batch) | OpenAI | 1050K | $1 | $5 | Reasoning, Vision |
| OpenAI: GPT-6.1 Sol | OpenAI | 1050K | $2 | $10 | Reasoning, Vision |
| OpenAI: GPT-6.1 Sol (batch) | OpenAI | 1050K | $1 | $5 | Reasoning, Vision |
| Anthropic: Claude Sonnet 5.5 | Anthropic | 1000K | $2 | $10 | Reasoning, Vision |
| Anthropic: Claude Sonnet 5.5 (batch) | Anthropic | 1000K | $1 | $5 | Reasoning, Vision |
| TypeSafe: Jev Router | Typesafe | 1000K | $0 | $0 | Reasoning, Vision |
| Perceptron: Perceptron Mk1.5 | Perceptron | 37K | $0.15 | $1.50 | Reasoning, Vision |
| Fireworks: Ember-1 | Fireworks | 1049K | $3 | $15 | Reasoning, Vision |
| Z.ai: GLM 5.3 Prime | Z-ai | 1000K | $2.80 | $8.80 | Reasoning |
| Qwen: Qwen3.8 Max Prime | Qwen | 1000K | $4 | $12 | Reasoning, Vision |
| AionLabs: Aion 3.5 Mini | Aion-labs | 262K | $0.70 | $1.40 | Reasoning |
| AionLabs: Aion 3.5 | Aion-labs | 262K | $3 | $6 | Reasoning |
| Upstage: Solar Mini 4 | Upstage | 524K | $0.05 | $0.20 | Reasoning |
| Cohere: Command A+ | Cohere | 192K | $0.30 | $1.50 | Reasoning, Vision |
| OpenAI: GPT-6 Luna Pro | OpenAI | 1050K | $0.10 | $0.50 | Reasoning, Vision |
| OpenAI: GPT-6 Luna Pro (batch) | OpenAI | 1050K | $0.05 | $0.25 | Reasoning, Vision |
| OpenAI: GPT-6 Luna | OpenAI | 1050K | $0.10 | $0.50 | Reasoning, Vision |
| OpenAI: GPT-6 Luna (batch) | OpenAI | 1050K | $0.05 | $0.25 | Reasoning, Vision |
| OpenAI: GPT-6 Sol Pro | OpenAI | 1050K | $2 | $10 | Reasoning, Vision |
| OpenAI: GPT-6 Sol Pro (batch) | OpenAI | 1050K | $1 | $5 | Reasoning, Vision |
| OpenAI: GPT-6 Sol | OpenAI | 1050K | $2 | $10 | Reasoning, Vision |
| OpenAI: GPT-6 Sol (batch) | OpenAI | 1050K | $1 | $5 | Reasoning, Vision |
| Anthropic: Claude Opus 5.5 | Anthropic | 1000K | $4 | $20 | Reasoning, Vision |
| Anthropic: Claude Opus 5.5 (batch) | Anthropic | 1000K | $2 | $10 | Reasoning, Vision |
| Xiaomi: MiMo-V2.6-Pro-UltraSpeed | Xiaomi | 1049K | $4.35 | $8.70 | Reasoning, Vision |
| Xiaomi: MiMo-V2.6-Flash | Xiaomi | 1049K | $0.14 | $0.28 | Reasoning, Vision |
| Xiaomi: MiMo-V2.6-Pro | Xiaomi | 1049K | $0.43 | $0.87 | Reasoning, Vision |
| SpaceXAI: Grok 4.7 | xAI | 500K | $2 | $6 | Reasoning, Vision |
| Qwen: Qwen3.8 Omni Flash | Qwen | 1000K | $0.15 | $0.47 | Reasoning, Vision |
| PrismML: Ternary Bonsai 2 27B | Prism-ml | 262K | $0.07 | $0.50 | Reasoning, Vision |
| Z.ai: GLM 5.3 FlashX | Z-ai | 1049K | $0.37 | $1.25 | Reasoning, Vision |
| DeepSeek: DeepSeek Pro Latest | ~deepseek | 1049K | $0.13 | $8 | Reasoning |
| DeepSeek: DeepSeek Flash Latest | ~deepseek | 1049K | $0.0089 | $0.80 | Reasoning, Vision |
| OpenAI: GPT Astra Latest | ~openai | 1050K | $10 | $50 | Reasoning, Vision |
| OpenAI: GPT Sol Latest | ~openai | 1050K | $2 | $10 | Reasoning, Vision |
| OpenAI: GPT Terra Latest | ~openai | 1050K | $2 | $12 | Reasoning, Vision |
| OpenAI: GPT Luna Latest | ~openai | 1050K | $0.10 | $0.50 | Reasoning, Vision |
Showing the top 50 of 321 models. Browse the full directory →
A reasoning model generates an internal chain of thought before producing its final answer, checking intermediate steps along the way. This makes it far more reliable at math, logic, planning, and debugging than a model that answers in one pass.
Use reasoning for problems with a verifiable right answer: math, code that must compile, data analysis, tricky logic. For casual chat, brainstorming, or writing, a standard model is faster and cheaper with no quality loss.
The thinking step generates additional tokens, so a reasoning response consumes more of your usage allowance than a standard reply from the same model. Many providers also price their dedicated reasoning models at a premium tier.
On CoreAI, reasoning-capable models currently come from Meta, NVIDIA, Liquid, Google, Stepfun and others. The list refreshes automatically as providers ship new thinking models.
Chat with GPT-5, Claude, Gemini and hundreds more — all in one app.