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. | — | 1B/4B/12B/27B |
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 | 128k |
| API pricingUSD per 1M tokens · lower wins | ||
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. | $0.05Relace | $0.04DeepInfra |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.2352DeepInfra | $0.05Parasail |
| $34.30DeepInfra · fp8 | $10.50Parasail · bf16 | |
| These models have no shared benchmark scores. | ||
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. | — | 3% |
DeepSWE 1.1Agentic coding — Artificial Analysis' independent test of deep, agentic software-engineering work — the AI has to plan and carry out substantial coding tasks end to end. (Version 1.1 of the test.) Higher is better. | 77% | — |
NL2Repo-BenchRepo-level code generation — Tests whether the AI can turn a natural-language requirement into working code across an entire repository, not just produce a single function or patch. Higher is better. | 68.2% | — |
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. | — | 29.7% |
HumanEvalFunction synthesis — 164 small Python problems: the AI is given a function's description and has to write the working function. This was the coding benchmark of the GPT-3.5 and GPT-4 era, before the field moved to fixing real bugs in real repositories. Higher is better. | — | 87.8% |
Terminal-Bench 4.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better. | 42.3% | — |
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. | 42.6% | — |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 42.4% |
GSM8KGrade-school math — Grade-school maths word problems that take a few steps of arithmetic to work through. It separated the models of 2022 and 2023 sharply, then saturated. One caveat on the historical numbers: OpenAI included part of the GSM8K training set in GPT-4's pre-training mix, so GPT-4's score is not a clean few-shot result. Higher is better. | — | 95.9% |
Agent's Last Exam · pass@1Agentic computer use — A hard set of desktop and operating-system tasks an AI agent has to finish by looking at the screen and working the machine itself. The score is the share it passes outright — partial credit does not count. Higher is better. | 35.4% | — |
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. | 60.5% | — |
Chartography · with toolsChart tasks — A chart-centred test run with tools available to the AI, reported separately from the chart-reading benchmarks above it. Higher is better. | 83% | — |
MMMUMultimodal — Tests the AI on understanding images and text together across many college subjects. Higher is better. | — | 64.9% |
| Overview | ||
| Company | DeepSeek | |
| Release date | Oct 17 2026 | Mar 12 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek-V4.1-Flash-1017vsClaude Haiku 5.5Gemma 3vsClaude Haiku 5.5DeepSeek-V4.1-Flash-1017vsGPT-6.1 SolGemma 3vsGPT-6.1 SolDeepSeek-V4.1-Flash-1017vsMuse Spark OpenGemma 3vsMuse Spark OpenDeepSeek-V4.1-Flash-1017vsGrok 4.7Gemma 3vsGrok 4.7DeepSeek-V4.1-Flash-1017vsMistral Large 4Gemma 3vsMistral Large 4DeepSeek-V4.1-Flash-1017vsKimi K3Gemma 3vsKimi K3Frequently asked questions
DeepSeek-V4.1-Flash-1017 and Gemma 3 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Gemma 3 shipped 584 days before DeepSeek-V4.1-Flash-1017, so benchmark comparisons should account for the intervening progress.
Context windows are 1M (DeepSeek-V4.1-Flash-1017) vs 128k (Gemma 3).
Direct benchmark comparisons are unavailable — DeepSeek-V4.1-Flash-1017 and Gemma 3 don't publish scores on any of the same benchmarks.