Best AI · Students
The best AI for students, from our test runs
On our study prompts (math with a checked answer, writing graded against a rubric, summaries and translations), DeepSeek V4 Pro and GPT-6 Astra each passed 20 of 20, the most; DeepSeek V4 Pro is first on the tie-break (the smaller message, then the lower cost). Here's every model's result, and which ones you can use free.
Test runs and prices checked . Updated .
Short answer
DeepSeek V4 Pro and GPT-6 Astra each passed 20 of 20, the most; DeepSeek V4 Pro is first on the tie-break (the smaller message, then the lower cost); the best everyday model was DeepSeek V4.1 Flash, 19 of 20. This is our own test, run on October 15, 2026: every prompt, reply and score is published.
Ranked by our test runs
This is our own test: the same prompts sent to every model through llmwise, each reply checked the same way, with every prompt, reply and score published. The study prompts test math, writing, summarization, and translation; they don't test a model's knowledge of any one course.
| # | Model | Passed | Hard ones | Free trial | On Pro | Cost per reply |
|---|---|---|---|---|---|---|
| 1 | DeepSeek V4 ProDeepSeek | 20 of 20 | 8 of 8 | No | 250 a month | $0.0012 |
| 2 | GPT-6 AstraOpenAI | 20 of 20 | 8 of 8 | No | 31 a month | $0.0108 |
| 3 | DeepSeek V4.1 FlashDeepSeek | 19 of 20 | 8 of 8 | No | 60 a day | $0.0004 |
| 4 | GPT-6.1 SolOpenAI | 19 of 20 | 8 of 8 | No | 125 a month | $0.0011 |
| 5 | Claude Haiku 4.5Anthropic | 19 of 20 | 7 of 8 | No | 250 a month | $0.0013 |
| 6 | Mistral Large 4Mistral | 19 of 20 | 7 of 8 | No | 250 a month | $0.0029 |
| 7 | Gemini 3.1 Pro (preview)Google | 19 of 20 | 7 of 8 | No | 125 a month | $0.0105 |
| 8 | Gemini 4 ArgonGoogle | 19 of 20 | 7 of 8 | No | 62 a month | $0.0014 |
| 9 | Claude Opus 5.5Anthropic | 19 of 20 | 7 of 8 | No | 62 a month | $0.0112 |
| 10 | GPT-6 LunaOpenAI | 18 of 20 | 8 of 8 | No | 60 a day | $0.0001 |
| 11 | GLM 5.3Z.ai | 18 of 20 | 7 of 8 | No | 250 a month | $0.0005 |
| 12 | Gemini 3.8 FlashGoogle | 18 of 20 | 7 of 8 | No | 250 a month | $0.0011 |
| 13 | Grok 4.7xAI | 18 of 20 | 7 of 8 | No | 250 a month | $0.0060 |
| 14 | Claude Sonnet 5.5Anthropic | 18 of 20 | 7 of 8 | No | 125 a month | $0.0034 |
| 15 | Claude Sonnet 5Anthropic | 17 of 20 | 8 of 8 | No | 125 a month | $0.0031 |
| 16 | GPT-6 SolOpenAI | 17 of 20 | 7 of 8 | No | 125 a month | $0.0023 |
| 17 | Kimi K3Moonshot | 17 of 20 | 7 of 8 | No | 125 a month | $0.0036 |
| 18 | Claude Haiku 5.5Anthropic | 16 of 20 | 7 of 8 | Yes | 60 a day | $0.0002 |
| 19 | Claude Fable 5.1Anthropic | 16 of 20 | 6 of 8 | No | 31 a month | $0.0156 |
| 20 | GLM 5.3 FlashZ.ai | 15 of 20 | 6 of 8 | No | 60 a day | $0.0002 |
The prompts, every reply and how each was scored: our test runs.
The best pick at each price
The same results, by what a message counts as on Pro: the best model at each size of message, then the rest of that size.
250 a month on Pro
DeepSeek V4 Pro, 20 of 20 passed; then Claude Haiku 4.5 19 of 20, Mistral Large 4 19 of 20, GLM 5.3 18 of 20, Gemini 3.8 Flash 18 of 20, and Grok 4.7 18 of 20.
31 a month on Pro
GPT-6 Astra, 20 of 20 passed; then Claude Fable 5.1 16 of 20.
60 a day on Pro (everyday models)
DeepSeek V4.1 Flash, 19 of 20 passed; then GPT-6 Luna 18 of 20, Claude Haiku 5.5 16 of 20, and GLM 5.3 Flash 15 of 20.
125 a month on Pro
GPT-6.1 Sol, 19 of 20 passed; then Gemini 3.1 Pro 19 of 20, Claude Sonnet 5.5 18 of 20, Claude Sonnet 5 17 of 20, GPT-6 Sol 17 of 20, and Kimi K3 17 of 20.
62 a month on Pro
Gemini 4 Argon, 19 of 20 passed; then Claude Opus 5.5 19 of 20.
Every limit is published. Paid plans also have a monthly fair-use limit on AI cost: Plus $3.75, Pro $7.50, Max $20, Ultra $42, Studio $85. Using every message on your plan at typical sizes stays under it; very large messages and heavy research use it faster. Every limit, explained.
Where the top models differ
The top 5, job by job.
| Job | DeepSeek V4 Pro | GPT-6 Astra | DeepSeek V4.1 Flash | GPT-6.1 Sol | Claude Haiku 4.5 |
|---|---|---|---|---|---|
| Math | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 |
| Writing | 5 of 5 | 5 of 5 | 4 of 5 | 4 of 5 | 4 of 5 |
| Summarization | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 |
| Translation | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 | 5 of 5 |
One prompt, side by side
An everyday math prompt, “Pens at 3 for $4”, where the models didn't all pass.
The prompt: Pens at 3 for $4 (everyday)
A shop sells pens at 3 for $4, or $1.50 for a single pen. What's the cheapest way to pay for exactly 10 pens, and what does it cost? Show your working briefly, then end with a line that says "Final answer: " followed by the answer alone.
GPT-6 Luna
Failed: Final answer 2 bundles and 4 singles for 12; expected 13.5.
For 10 pens, compare buying 0–3 bundles of 3: - 0 bundles: 10 singles = $15 - 1 bundle: 7 singles = $14.50 - 2 bundles: 4 singles = $12 - 3 bundles: 1 single = $13.50 Final answer: 2 bundles and 4 singles for $12
402 tokens in, 198 out (111 of them reasoning) · 2.9 s · $0.0001 · 1 message on Pro · answered by openai/gpt-6-luna via OpenAI ·
Claude Sonnet 5
Passed: Final answer Buy 3 packs of 3 pens ( 4 each) plus 1 single pen ( 1.50), total 13.50.: right.
Compare unit costs: 3-for-$4 = $1.33/pen, cheaper than the $1.50 single price. So maximize the number of 3-packs. 10 pens = 3 packs of 3 (9 pens) + 1 single pen Cost = 3×$4 + 1×$1.50 = $12 + $1.50 = $13.50 Check alternative (2 packs + 4 singles): 2×$4 + 4×$1.50 = $8 + $6 = $14 (worse) Check buying an extra pack for surplus: 4 packs = $16 for 12 pens (worse, and you don't need 12) Final answer: Buy 3 packs of 3 pens ($4 each) plus 1 single pen ($1.50), total $13.50.
545 tokens in, 505 out (250 of them reasoning) · 6.3 s · $0.0061 · 1 message on Pro · answered by anthropic/claude-sonnet-5 via Claude Platform on AWS ·
Kimi K3
Passed: Final answer Buy three 3-for- 4 packs and one single pen; 13.50: right.
Three 3-for-$4 packs give 9 pens for $12; add 1 single for $1.50 to reach exactly 10. Total = $12 + $1.50 = $13.50. Final answer: Buy three 3-for-$4 packs and one single pen; $13.50
489 tokens in, 322 out (247 of them reasoning) · 5.4 s · $0.0046 · 1 message on Pro · answered by moonshotai/kimi-k3 via Phala ·
How we ranked them
Ranked by how many prompts each model passed, then how many of the hard ones, then by the smaller message (the everyday models first), then by the lower cost per reply. No ranking is chosen by hand.
- Our test runs: 50 prompts on every model
- The AI model leaderboard, job by job
- Best AI assistants, tested (October 2026)
- The best AI for research, from our test runs
- The best free AI, from our test runs
- The best AI model right now: all 20, ranked by our tests
- The best AI for business, from our test runs
- The best AI for data analysis, from our test runs
- AI chat with every top model in one place
Questions
What is the best AI for students?
In our own test runs on October 2026, on the study prompts: DeepSeek V4 Pro and GPT-6 Astra each passed 20 of 20, the most; DeepSeek V4 Pro is first on the tie-break (the smaller message, then the lower cost), and DeepSeek V4.1 Flash did best of the everyday models (19 of 20).
Is there a free AI for students?
To try: 5 free messages on sign-up, once, on Claude Haiku 5.5, with no card. After that, Pro is $20 a month for every model.
Can I use AI for homework?
Check your school's rules first: many allow AI for explaining and practice but not for graded work. A model that shows its steps, as the math prompts here ask, is easier to learn from than a bare answer.
What is the best AI for homework?
For the homework we test: on math with a checked final answer, Claude Haiku 5.5, GLM 5.3 Flash, DeepSeek V4.1 Flash, GLM 5.3, DeepSeek V4 Pro, Gemini 3.8 Flash, Claude Haiku 4.5, Mistral Large 4, Grok 4.7, GPT-6.1 Sol, GPT-6 Sol, Claude Sonnet 5.5, Claude Sonnet 5, Kimi K3, Gemini 3.1 Pro, Gemini 4 Argon, Claude Opus 5.5, GPT-6 Astra, and Claude Fable 5.1 each passed 5 of 5, the most; Claude Haiku 5.5 is first on the tie-break (the smaller message, then the lower cost); on writing graded against a rubric, GPT-6 Luna, DeepSeek V4 Pro, and GPT-6 Astra each passed 5 of 5, the most; GPT-6 Luna is first on the tie-break (the smaller message, then the lower cost). Our free AI homework helper (1 free run a day on Claude Haiku 5.5, no account) explains the steps before the answer.
Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.
See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.