Compare AI models
| Specifications | ||
ParametersThe learned values that shape how a model responds. B means billion and T means trillion. For mixture-of-experts models, the total count is larger than the number active for each token. Parameter count alone does not measure quality. | 1.6T | — |
Context windowHow much information the model can work with in one request, measured in tokens — pieces of words or other data. This includes your question, attachments, conversation and reply. Limits can vary by model variant and provider. | 1M | 1M |
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.66 | $0.25 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $1.98 | $1.50 |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | $0.022 | $0.025 |
Cheapest inputLowest input rate across every provider on OpenRouter, the lab included. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.0999Relace | $0.125Google AI Studio |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $1.98DeepSeek | $0.75Google AI Studio |
| $231DeepSeek list price | $125Google list price | |
| Benchmarks | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | 35% | 11% |
Terminal-Bench 2.1Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Higher is better. | 87.9% | 31% |
Humanity's Last Exam · no toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects, written so you can't just look up the answer. “No tools” means the AI answers on its own. Higher is better. | 42.7% | 16% |
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | — | 38.3% |
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 80.6% | — |
MLE-BenchML engineering — Can the AI do the work of a machine-learning engineer? It competes in real Kaggle competitions — building, training, and tuning models end to end — and the score reflects how well it places. Higher is better. | — | 22% |
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better. | — | 72% |
Toolathlon-VerifiedPersonal tool use — Tests how well the AI uses everyday personal tools and apps to get things done — a human-checked version of Toolathlon. Higher is better. | 74.1% | — |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | 83.4% | — |
CyberGymCybersecurity — Tests the AI on cybersecurity challenges — finding and exploiting software weaknesses inside a safe sandbox. Higher is better. | 83.3% | — |
Humanity's Last Exam · with toolsMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | 60% | — |
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | 61.3% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 86.9% |
SimpleQA VerifiedKnowledge — accuracy on the vetted factual questions in SimpleQA Verified | — | 43.3% |
MMMLUKnowledge — average accuracy on multilingual translations of MMLU | — | 88.9% |
OSWorld-VerifiedAgentic computer use — Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better. | — | 54.3% |
AutomationBenchBusiness workflows — Tests whether the AI can run real multi-step business workflows — the kind of end-to-end office processes companies want to automate — from start to finish. Higher is better. | 31.8% | — |
GDPval-AA v2Knowledge work — economically valuable knowledge work (v2, re-based Elo) | — | 642 |
CharXiv ReasoningChart reasoning — Can the AI read and reason about complex charts and figures, not just text? Higher is better. | — | 73.2% |
CharXiv Reasoning · with toolsChart reasoning — The same chart-and-figure reasoning test, run with the AI allowed to use tools — writing code to inspect the image, for instance — rather than reading the chart unaided. Scores run higher than the unaided version, so read the two as separate tests. Higher is better. | — | 75.6% |
MMMU-ProMultimodal reasoning — A tougher version of MMMU — college-level questions that mix images, diagrams, and text together. Higher is better. | — | 76.8% |
MRCR v2 (8-needle) · 128k averageLong context — Tests whether the AI can find specific details buried inside a very long document (around 128k tokens — roughly a long book). Higher is better. | — | 60.1% |
MRCR v2 (8-needle) · 1M pointwiseLong context — Tests whether the AI can find specific details buried inside an enormous document (around 1 million tokens — many books). Higher is better. | — | 12.3% |
| Overview | ||
| Company | DeepSeek | |
| Release date | Aug 13 2026 | Mar 3 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
DeepSeek-V4-Pro-0813vsClaude Fable 5.2Gemini 3.1 Flash-LitevsClaude Fable 5.2DeepSeek-V4-Pro-0813vsGPT-6.1 TerraGemini 3.1 Flash-LitevsGPT-6.1 TerraDeepSeek-V4-Pro-0813vsMuse EmberGemini 3.1 Flash-LitevsMuse EmberDeepSeek-V4-Pro-0813vsGrok 4.7Gemini 3.1 Flash-LitevsGrok 4.7DeepSeek-V4-Pro-0813vsMistral Large 4Gemini 3.1 Flash-LitevsMistral Large 4DeepSeek-V4-Pro-0813vsKimi K3.1 ProGemini 3.1 Flash-LitevsKimi K3.1 ProFrequently asked questions
DeepSeek-V4-Pro-0813 leads Gemini 3.1 Flash-Lite on 3 of the 3 benchmarks they both report (BullshitBench v2, Terminal-Bench 2.1, Humanity's Last Exam). Gemini 3.1 Flash-Lite is cheaper on both input and output: $0.25 vs $0.66 per million input tokens, and $1.50 vs $1.98 per million output tokens. Figures are base-tier rates. Gemini 3.1 Flash-Lite shipped 163 days before DeepSeek-V4-Pro-0813, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (DeepSeek-V4-Pro-0813) vs 1M (Gemini 3.1 Flash-Lite). DeepSeek-V4-Pro-0813 is open weight, while Gemini 3.1 Flash-Lite is closed.
On BullshitBench v2, DeepSeek-V4-Pro-0813 leads at 35% vs Gemini 3.1 Flash-Lite at 11%. On Terminal-Bench 2.1, DeepSeek-V4-Pro-0813 leads at 87.9% vs Gemini 3.1 Flash-Lite at 31%. On Humanity's Last Exam · no tools, DeepSeek-V4-Pro-0813 leads at 42.7% vs Gemini 3.1 Flash-Lite at 16%.