Views
No views yet
"We didn't build another coding assistant. We built the engineer that engineers wish they had."
| Model | HumanEval+ | MBPP+ | Multi-File Refactor | Architecture Design | Latency | Verdict |
|---|---|---|---|---|---|---|
| 🥇 List-3.0-Ultra-Coder | 98.2% | 97.8% | 96.5% | 97.1% | 38ms | 👑 King |
| Claude Opus 4.7 | 97.8% | 97.2% | 95.8% | 96.4% | 1200ms | Titan |
| Gemini 3.1 Ultra | 97.5% | 97.0% | 94.2% | 95.8% | 850ms | Titan |
| GPT-5.4 Pro | 95.1% | 94.8% | 91.3% | 93.2% | 900ms | |
| DeepSeek-V3 | 94.8% | 94.5% | 90.7% | 92.1% | 400ms | |
| Llama 4-405B | 94.2% | 94.0% | 89.5% | 91.8% | 600ms | |
| Qwen3-235B-A22B | 93.8% | 93.5% | 88.9% | 90.5% | 350ms | |
| Mistral Large 3 | 93.2% | 93.0% | 87.3% | 89.7% | 300ms |
38ms average latency. That's not a typo. Our MoE routing activates only 8 of 256 experts per token, giving you the intelligence of a 228B model with the speed of a 7B model.
| Feature | List-2.0 | List-3.0 |
|---|---|---|
| Parameters | 500B (Dense) | 228B (MoE) |
| Active Parameters | 500B | ~7B per token |
| Expert Networks | — | 256 Specialists |
| Context Window | 128K | 204,800 tokens |
| Multi-Token Prediction | ⌠| ✅ 3-token lookahead |
| FP8 Quantization | ⌠| ✅ Dynamic |
| Speed vs 2.0 | 1x | ~31x faster |
| Architecture Reasoning | Good | State-of-the-art |
| Security Auditing | Basic | Enterprise-grade |
1Architecture: Mixture-of-Experts (MoE) with Multi-Token Prediction (MTP)
2Total Parameters: 228,000,000,000 (228B)
3Active per Token: ~7B (8 of 256 experts)
4Expert Networks: 256 specialized routing experts
5MTP Modules: 3 (predicts 3 tokens ahead simultaneously)
6Hidden Size: 3,072
7Attention Heads: 48 (8 KV heads, GQA)
8Layers: 62 transformer blocks
9Context Window: 204,800 tokens (~400 pages of code)
10Quantization: FP8 (float8_e4m3fn) with dynamic activation
11Precision: BFloat16 (training) / FP8 (inference)
12Vocabulary: 200,064 tokens
13RoPE θ: 5,000,000 (extreme long-context support)💡 The IDE provides native integration with all List models, including real-time code completion, multi-file refactoring, and architectural guidance.
1from transformers import AutoModelForCausalLM, AutoTokenizer
2
3model_name = "List-cloud/List-3.0-Ultra-Coder-Brain"
4tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
5model = AutoModelForCausalLM.from_pretrained(
6 model_name,
7 device_map="auto",
8 trust_remote_code=True,
9 torch_dtype="auto"
10)
11
12prompt = "Implement a lock-free concurrent hash map in Rust with work-stealing."
13inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
14outputs = model.generate(**inputs, max_new_tokens=4096)
15print(tokenizer.decode(outputs[0], skip_special_tokens=True))âš ï¸ Local deployment requires 8x A100 80GB or equivalent. For most users, the API or IDE is recommended.
| Domain | Capability |
|---|---|
| ðŸ—ï¸ Architecture Design | Design entire system architectures from a single prompt. Microservices, event-driven, CQRS — it knows them all. |
| 🔄 Multi-File Refactoring | Understands 200K+ tokens of context. Refactor across hundreds of files with full dependency awareness. |
| 🔒 Security Auditing | Identifies OWASP Top 10, supply chain vulnerabilities, and zero-day patterns in real-time. |
| 🧪 Test Generation | Generates comprehensive test suites with edge cases, mocks, and integration tests. |
| 📚 Documentation | Produces production-ready docs, API references, and architecture decision records (ADRs). |
| 🛠Debugging | Traces bugs across stack traces, async boundaries, and distributed systems. |
| Product | Description |
|---|---|
| List Coder IDE | Full-featured code editor with native AI integration |
| List-1.0-Ultra-Coder | Fast, lightweight model for everyday coding |
| List-2.0-Ultra-Coder | High-performance dense model for complex tasks |
| List-3.0-Ultra-Coder | Our flagship — 228B MoE powerhouse |
| List-Stack-10M | Specialized for full-stack web development |