LaunchesDeepSeekToday
DeepSeek
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
Running this yourself: likely needs a high-memory cloud gpu.
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions.
OpenRouter
Price record: 2026-10-03. Source: openrouter.
No current subscription pricing is tracked for this model.
Confirm this specific model, usage limits, and billing terms with the provider. A subscription does not automatically include API credits.
No login needed to compare. Prices are in USD; provider charges are separate from AI Market Cap plans. Context length, caching, tools, taxes, and regional terms can change the final cost. Open weights do not mean free hosting.
61.4
Quality Score
1334
Arena ELO
671B
Parameters
164K
Context
This measures the amount of verifiable public evidence we have, not how capable the model is. A missing field means it has not been verified yet, not that its value is zero.
17 of 22 public signals
Sign in to join the discussion
0
Downloads
0
Likes
Dec 2024
Released
4/5 signals
4/4 signals
4/5 signals
1/4 signals
4/4 signals
Parameters
671B
Training compute
3.3e24 FLOP
Dataset scale
14.8T units
Base model
Not reported
Source-reported access: Open weights (restricted use) · Confident confidence
Gaps we are still tracking
Launches
1
Pricing
1
Benchmarks
17
Open Source
1
Research
3
General
2
Recent launch, pricing, benchmark, and API signals linked to this model or its provider.
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
View sourceSWE-Bench Verified resolved rate 60.0
View sourceLiveCodeBench pass@1 49.6 across 1055 tasks
View sourceGenerating Scalable Vector Graphics (SVG) code from natural-language instructions is an open-ended task without absolute visual ground truth, leaving both evaluation and policy optimization without a faithful signal. Scalar metrics (CLIP, Aesthetic) calibrated on natural images transfer poorly to stylized vector content, and reusing them as RL rewards triggers reward hacking. We address both limitations with rubric-based scoring. We first establish empirically that prompting a vision-language judge with a multi-axis rubric correlates with human judgments far better than scalar metrics, both across samples and within instructions. Building on this finding, we introduce RULER (Instance-aware Rubric Rewards for Reinforcement Learning), which converts each instruction into an instance-aware rubric of six items spanning semantic, visual, and stylistic axes; a judge VLM scores rendered rollouts item-by-item, and the weighted satisfactions form a fine-grained reward optimized via Group Relative Policy Optimization. Because the rubric is derived from text alone, RULER requires neither paired SVG ground truth nor human preference labels. On MMSVG-Illustration and MMSVG-Icon, RULER lifts the rubric score from 0.432/0.395 to 0.693/0.683, surpassing dedicated SVG specialists and matching the substantially larger DeepSeek-V3, with ablations identifying rubric design as the active lever for RL on open-ended SVG generation. The project page is available at https://hangyuran.github.io/RULER/.
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.
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 is now available through local Ollama runtime. 160K context window listed. A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
SWE-Bench Verified resolved rate 60.0
LiveCodeBench pass@1 49.6 across 1055 tasks
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6
language: 0.3 | coding: 0.3 | instruction_following: 1.0 | Overall: 0.6