Best model for your job
Best AI assistant for everyday use
For everyday use, GPT-6 Astra scores highest on the current data (77.5), weighting writing & preference 30%, knowledge 30%, reasoning 20%, instruction following 20%. The cheapest model in the top 10 is Gemini 3.8 Flash at $0.75 / $3.75 per million tokens.
Last verified
Weights: Writing & Preference 30%, Knowledge 30%, Reasoning 20%, Instruction Following 20%. A general assistant needs to be good at a bit of everything.
| # | Model | Provider | Fit | Writing & Preference | Knowledge | Reasoning | Instruction Following | Input $/M | Output $/M |
|---|---|---|---|---|---|---|---|---|---|
| 1 | GPT-6 Astra | OpenAI | 77.5 | 75.3 | 75.3 | 85.1 | 76.3 | $10 | $50 |
| 2 | Claude Fable 5.1 | Anthropic | 75.8 | 79.2 | 69.6 | 76.7 | 79.2 | $10 | $50 |
| 3 | Claude Opus 5.5 | Anthropic | 75.4 | 78.2 | 66.4 | 80.2 | 80.0 | $4 | $20 |
| 4 | Gemini 3.8 Flash | 75.1 | 72.2 | 74.8 | 76.9 | 78.0 | $0.75 | $3.75 | |
| 5 | Claude Opus 5 | Anthropic | 75.1 | 79.2 | 66.8 | 77.2 | 79.2 | $5 | $25 |
| 6 | Claude Fable 5 | Anthropic | 72.5 | 75.9 | 62.2 | 76.8 | 78.6 | $10 | $50 |
| 7 | GPT-6.1 Sol | OpenAI | 72.4 | 63.6 | 71.8 | 81.9 | 77.0 | $2 | $10 |
| 8 | Gemini 3.7 Flash | 71.8 | 71.2 | 69.7 | 70.0 | 77.7 | $0.75 | $3.75 | |
| 9 | GPT-5.6 Sol | OpenAI | 71.8 | 73.3 | 64.3 | 74.8 | 77.7 | $4 | $20 |
| 10 | GPT-5.5 | OpenAI | 71.2 | 72.7 | 64.4 | 72.8 | 77.5 | $5 | $30 |
| 11 | Gemini 3.1 Pro Preview | 71.1 | 66.1 | 71.8 | 71.7 | 77.0 | $2 | $12 | |
| 12 | GPT-6 Sol | OpenAI | 70.7 | 71.9 | 64.8 | 74.0 | 74.5 | $2 | $10 |
| 13 | Kimi K3 (open weights) | Moonshot AI | 70.1 | 76.6 | 63.2 | 63.0 | 77.7 | $3 | $15 |
| 14 | GPT-5.4 | OpenAI | 69.0 | 71.9 | 65.3 | 61.8 | 77.1 | $2.50 | $15 |
| 15 | Claude Opus 4.8 | Anthropic | 68.4 | 72.0 | 61.3 | 64.7 | 77.4 | $5 | $25 |
| 16 | Claude Opus 4.6 | Anthropic | 68.1 | 73.5 | 61.9 | 57.8 | 79.5 | $5 | $25 |
| 17 | Gemini 3.6 Flash | 68.0 | 68.2 | 67.8 | 58.8 | 77.0 | $0.75 | $3.75 | |
| 18 | Claude Opus 4.7 | Anthropic | 67.8 | 75.1 | 62.6 | 53.8 | 78.4 | $5 | $25 |
| 19 | Gemini 3.5 Flash | 67.5 | 65.5 | 66.3 | 62.8 | 77.0 | $1.50 | $9 | |
| 20 | GPT-5.6 Terra | OpenAI | 66.8 | 70.2 | 61.2 | 60.7 | 76.4 | $2 | $12 |
| 21 | Claude Sonnet 5.5 | Anthropic | 66.0 | 66.0 | 66.0 | 54.0 | 78.3 | $2 | $10 |
| 22 | Grok 4.6 | xAI | 65.0 | 62.3 | 63.3 | 61.4 | 75.4 | $2 | $6 |
| 23 | Qwen3.8 Max | Alibaba (Qwen) | 65.0 | 67.1 | 61.7 | 54.4 | 77.6 | $2 | $6 |
| 24 | Gemini 3 Pro | 65.0 | 66.4 | 64.4 | 52.5 | 76.3 | — | — | |
| 25 | GLM-5.3 (open weights) | Z.ai (Zhipu) | 64.9 | 75.7 | 58.3 | 46.1 | 77.5 | $1.40 | $4.40 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Top two head to head: GPT-6 Astra vs Claude Fable 5.1
Frequently asked questions
What is the best ai assistant for everyday use?
For everyday use, GPT-6 Astra scores highest on the current data (77.5), weighting writing & preference 30%, knowledge 30%, reasoning 20%, instruction following 20%. The cheapest model in the top 10 is Gemini 3.8 Flash at $0.75 / $3.75 per million tokens.
How is this shortlist built?
A general assistant needs to be good at a bit of everything. Each model's category scores are blended with those weights; only ranked models with results in every needed category are listed.
Other jobs
- Best AI model for coding
- Best AI model for debugging
- Best AI model for building websites
- Best local LLM for coding
- Best AI model for agents
- Best AI model for RAG
- Best AI model for research
- Best AI model for studying
- Best AI model for math
- Best AI model for hard reasoning
- Best AI model for data analysis
- Best AI model for Excel
- Best AI model for data extraction
- Best AI model for writing
- Best AI model for emails
- Best AI model for copywriting
- Best AI model for creative writing
- Best AI model for translation
- Best AI model for language learning
- Best AI model for customer service
- Best AI model for presentations
- Best AI model for PDFs