DeepSeek-V3

Models & Pricing

The Temperature Parameter

⚡️ Efficiency Gains 🤖 DSA achieves fine-grained sparse attention with minimal impact on output quality — boosting long-context performance & reducing compute cost. 📊 Benchmarks show V3.2-Exp perform

⚡️ Efficiency Gains 🤖 DSA achieves fine-grained sparse attention with minimal impact on output quality — boosting long-context performance & reducing compute cost. 📊 Benchmarks show V3.2-Exp performs on par with V3.1-Terminus. 2/n https://t.co/zTG679p5Zm

Introducing DeepSeek-V3.1: our first step toward the agent era! 🚀 🧠 Hybrid inference: Think & Non-Think — one model, two modes ⚡️ Faster thinking: DeepSeek-V3.1-Think reaches answers in less time vs

Introducing DeepSeek-V3.1: our first step toward the agent era! 🚀 🧠 Hybrid inference: Think & Non-Think — one model, two modes ⚡️ Faster thinking: DeepSeek-V3.1-Think reaches answers in less time vs. DeepSeek-R1-0528 🛠️ Stronger agent skills: Post-training boosts tool use and

Context Caching is Available 2024/08/02

StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling

Long-horizon agents can fail even when their underlying models can solve the constituent steps. They may lose track of mutable state, fail to reactivate lessons from earlier executions, skip known procedures, or stop prematurely. We bet on harness scaling to improve the execution system around an agent without changing its model weights. We introduce StateM, an agent-native runtime that organizes execution around durable states, phase-local context, checked transitions, recoverable runbooks, and versioned procedural practices that agents and users can inspect together. On Terminal-Bench 2.1, StateM raises GPT-5.5 xhigh to 92.1\%, versus 83.1\% reference and GPT-5.6 Sol Ultra at 91.9\%. The runbook transfers unchanged to GPT-5.6. With GPT-5.6 Sol xhigh, StateM reaches 95.3\% raw accuracy across 445 trials and succeeds on all 89 tasks at least once. The frozen profile raises GPT-5.6 Luna from 76.7 to 85.4\%, above the 84.9\% Sol xhigh reference. Using the same runtime, runbook structure, and golden rules, less than \38 of adaptation raises DeepSeek-V4 Flash from 82.7 to 88.1% under standard timeouts and to 89.1% on an 88-task common core. Extending only the remaining latency-sensitive task matches the reported 88.8% GPT-5.6 Sol max result. Final-score API usage is about 15 versus \574.68 for the GPT reference; total DeepSeek expenditure is 52.22. On BusinessBench, family-specific runbooks built on development sets yield held-out gains of 0.55 macro and 1.34 micro points; two mechanism-matched families improve by 10.04 points. Concrete rules generalize when tasks share execution structure, while the control methodology applies broadly. StateM turns selected postmortem findings into persistent, executable preconditions and practices, making learned controls explicit and enforceable through stateful controls. Code at github.com/henryqin1997/statem.

Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation

We introduce Nemotron-Cascade 2, an open 30B MoE model with 3B activated parameters that delivers best-in-class reasoning and strong agentic capabilities. Despite its compact size, its mathematical and coding reasoning performance approaches that of frontier open models. It is the second open-weight LLM, after DeepSeekV3.2-Speciale-671B-A37B, to achieve Gold Medal-level performance in the 2025 International Mathematical Olympiad (IMO), the International Olympiad in Informatics (IOI), and the ICPC World Finals, demonstrating remarkably high intelligence density with 20x fewer parameters. In contrast to Nemotron-Cascade 1, the key technical advancements are as follows. After SFT on a meticulously curated dataset, we substantially expand Cascade RL to cover a much broader spectrum of reasoning and agentic domains. Furthermore, we introduce multi-domain on-policy distillation from the strongest intermediate teacher models for each domain throughout the Cascade RL process, allowing us to efficiently recover benchmark regressions and sustain strong performance gains along the way. We release the collection of model checkpoint and training data.

Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillation

We introduce Nemotron-Cascade 2, an open 30B MoE model with 3B activated parameters that delivers best-in-class reasoning and strong agentic capabilities. Despite its compact size, its mathematical and coding reasoning performance approaches that of frontier open models. It is the second open-weight LLM, after DeepSeekV3.2-Speciale-671B-A37B, to achieve Gold Medal-level performance in the 2025 International Mathematical Olympiad (IMO), the International Olympiad in Informatics (IOI), and the ICPC World Finals, demonstrating remarkably high intelligence density with 20x fewer parameters. In contrast to Nemotron-Cascade 1, the key technical advancements are as follows. After SFT on a meticulously curated dataset, we substantially expand Cascade RL to cover a much broader spectrum of reasoning and agentic domains. Furthermore, we introduce multi-domain on-policy distillation from the strongest intermediate teacher models for each domain throughout the Cascade RL process, allowing us to efficiently recover benchmark regressions and sustain strong performance gains along the way. We release the collection of model checkpoint and training data.

deepseek-v3.2-reasoner - SWE-Bench Verified

SWE-Bench Verified resolved rate 60.0

DeepSeek-V3 - LiveCodeBench

LiveCodeBench pass@1 49.6 across 1055 tasks

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6

deepseek-ai/DeepSeek-V4-Pro-0813 · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Pro-0813.

deepseek-ai/DeepSeek-V4-Pro-DSpark · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Pro-DSpark.

deepseek-ai/DeepSeek-V4-Pro · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Pro.

deepseek-ai/DeepSeek-V4-Flash-DSpark · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Flash-DSpark.

deepseek-ai/DeepSeek-V4-Flash · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Flash.

deepseek-ai/DeepSeek-V4-Flash-0731 · Hugging Face

DeepSeek published benchmark or leaderboard evidence for DeepSeek-V4-Flash-0731.

deepseek-v3 — LiveBench Scores

language: 0.5 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6