Dari unggahan peluncuran Anthropic, yang memberi angka benchmark dalam teks dan tidak menyebut jendela konteks, output maksimum, atau angka harga. · sumber
Estimate from the version's published specs, benchmarks and reviews, 0-5; scores are only comparable within one criteria year. Scored as launched (upgraded Claude 3.5 Sonnet, 22 Oct 2024): 200K-token context (OpenRouter model page, released 22 Oct 2024; the Anthropic post restates no window); price unchanged at $3 in / $15 out = $6 blended at 3:1 (Willison, Zvi); text and image input plus computer-use beta. Anthropic reports SWE-bench Verified 49.0% (33.4% before) and OSWorld 14.9% screenshot-only; neither maps to an anchor. Reasoning null (no launch-version AA index opened), coding null (no Terminal-Bench 4.0), speed null (no tokens/s published). Reviews: Willison (hands-on with the computer-use demo), Zvi, TechCrunch, The Decoder.[1]
It worked exactly as advertised.
We only have very early feedback so far, so it's hard to tell how much what I will be calling Claude 3.5.1 improves performance in practice over Claude 3.5. It does seem like it is a clear improvement.
But don't fire your secretary just yet. In an evaluation designed to test an AI agent's ability to help with airline booking tasks, like modifying a flight reservation, the new 3.5 Sonnet managed to complete less than half of the tasks successfully.
The new Claude 3.5 Sonnet model shows improved performance, especially in logical reasoning, mathematical problem-solving and programming tasks.
| Initial explorations of Anthropic's new Computer Use capability · 22 Okt 2024 | 0.60 |
|---|---|
| Anthropic's new AI model can control your PC · 22 Okt 2024 | -0.20 |
| Anthropic releases updated, smarter Claude Haiku 3.5 and Sonnet 3.5 model · 22 Okt 2024 | 0.70 |
| Claude Sonnet 3.5.1 and Haiku 3.5 · 24 Okt 2024 | 0.30 |
Foto dari pembaca, terpisah dari media bersumber milik pin. Nilai apa yang Anda makan atau lihat, dan di pin restoran, tandai hidangan langsung pada foto Anda.
Entri pertama selalu merupakan sumber pin. Keyakinan keseluruhan adalah rata-rata tertimbang dari seberapa kuat setiap referensi mendukung waktu mulai dan berakhir yang digunakan di atas; bobot sebuah referensi berkurang setengahnya untuk setiap 180 hari lebih lama dari yang terbaru.
Unggahan itu bertanggal “22 Okt 2024” dan menyatakan “Claude 3.5 Sonnet yang ditingkatkan kini tersedia untuk semua pengguna. Mulai hari ini, pengembang dapat membangun dengan beta penggunaan komputer”; TechCrunch[4] dan CNBC[5] melaporkan rilis itu “pada hari Selasa” yang sama.
Anthropic's[1][3] model card addendum: the upgraded 3.5 Sonnet sets new state-of-the-art results in agentic coding (SWE-bench Verified), agentic tasks (TAU-bench) and computer use from screenshots (OSWorld), with a 14.9% average OSWorld success rate from screenshots alone.
Anthropic's[1][2] research post on the skill: Claude looks at screenshots and "counts how many pixels vertically or horizontally it needs to move a cursor in order to click in the correct place", and was trained on simple software such as a calculator and a text editor without internet access.
TechCrunch, 2024-10-22: "Anthropic[1][2][3] on Tuesday released an upgraded version of its Claude 3.5 Sonnet model that can understand and interact with any desktop app" through a Computer Use API in open beta on the API, Bedrock and Vertex AI.
CNBC, 2024-10-22: chief science officer Jared Kaplan says it can do tasks with "tens or even hundreds of steps"; Amazon had early access and beta testers included Asana, Canva and Notion; released Tuesday in public beta for developers.
Ada yang kurang atau salah? Sampaikan dengan kata-kata Anda sendiri: tautan yang mendukung pin ini, tanggal mulai atau berakhir yang berbeda beserta alasannya, atau fakta yang kurang atau keliru. AI memeriksanya terhadap sumber pin ini, mencari sumber yang lebih baik, dan menambahkan halaman mana pun yang mendukung Anda. Sumber pin itu sendiri tetap paling diperhitungkan. Gambar yang menampilkan hal lain, atau menampilkannya dengan buruk, juga diperiksa, lalu diturunkan atau diganti.
Judul asli: Claude 3.5 Sonnet Upgrade Released
| Tolok ukur | Claude 3.5 Sonnet (new) | Claude 3.5 Haiku | Claude 3.5 Sonnet | GPT-4o* | GPT-4o mini* | Gemini 1.5 Pro | Gemini 1.5 Flash |
|---|---|---|---|---|---|---|---|
| Penalaran tingkat pascasarjana (GPQA Diamond) | 65.0% (0-shot CoT) | 41.6% (0-shot CoT) | 59.4% (0-shot CoT) | 53.6% (0-shot CoT) | 40.2% (0-shot CoT) | 59.1% (0-shot CoT) | 51.0% (0-shot CoT) |
| Pengetahuan tingkat sarjana (MMLU Pro) | 78.0% (0-shot CoT) | 65.0% (0-shot CoT) | 75.1% (0-shot CoT) | — | — | 75.8% (0-shot CoT) | 67.3% (0-shot CoT) |
| Kode (HumanEval) | 93.7% (0-shot) | 88.1% (0-shot) | 92.0% (0-shot) | 90.2% (0-shot) | 87.2% (0-shot) | — | — |
| Pemecahan soal matematika (MATH) | 78.3% (0-shot CoT) | 69.2% (0-shot CoT) | 71.1% (0-shot CoT) | 76.6% (0-shot CoT) | 70.2% (0-shot CoT) | 86.5% (4-shot CoT) | 77.9% (4-shot CoT) |
| Kompetisi matematika SMA (AIME 2024) | 16.0% (0-shot CoT) | 5.3% (0-shot CoT) | 9.6% (0-shot CoT) | 9.3% (0-shot CoT) | — | — | — |
| Tanya jawab visual (MMMU) | 70.4% (0-shot CoT) | — | 68.3% (0-shot CoT) | 69.1% (0-shot CoT) | 59.4% (0-shot CoT) | 65.9% (0-shot CoT) | 62.3% (0-shot CoT) |
| Pemrograman agen (SWE-bench Verified) | 49.0% | 40.6% | 33.4% | — | — | — | — |
| Penggunaan alat agen (TAU-bench), ritel | 69.2% | 51.0% | 62.6% | — | — | — | — |
| Penggunaan alat agen (TAU-bench), maskapai | 46.0% | 22.8% | 36.0% | — | — | — | — |
* Tabel evaluasi Anthropic tidak menyertakan keluarga model o1 milik OpenAI, karena bergantung pada waktu komputasi pra-respons yang ekstensif, tidak seperti model pada umumnya.