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 | 123B |
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 | 256k |
| 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 | — |
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 | — |
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 | — |
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. | $0.12Relace | $0.40Mistral |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $0.40Baidu | $2.00Mistral |
| $46.66Baidu · fp8 | $180Mistral | |
| Benchmarks | ||
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% | 72.2% |
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% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | — | 61.3% |
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% | — |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | — | 32.6% |
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 · 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% | — |
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% | — |
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% | — |
| Overview | ||
| Company | DeepSeek | Mistral |
| Release date | Aug 13 2026 | Dec 9 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
DeepSeek-V4-Pro-0813vsClaude Haiku 5.5Devstral 2 (123B)vsClaude Haiku 5.5DeepSeek-V4-Pro-0813vsGPT-6.1 SolDevstral 2 (123B)vsGPT-6.1 SolDeepSeek-V4-Pro-0813vsGemini 4 ArgonDevstral 2 (123B)vsGemini 4 ArgonDeepSeek-V4-Pro-0813vsMuse Spark OpenDevstral 2 (123B)vsMuse Spark OpenDeepSeek-V4-Pro-0813vsGrok 4.7Devstral 2 (123B)vsGrok 4.7DeepSeek-V4-Pro-0813vsKimi K3Devstral 2 (123B)vsKimi K3Frequently asked questions
DeepSeek-V4-Pro-0813 leads Devstral 2 (123B) on 1 of the 1 benchmark they both report (SWE-Bench Verified). Only DeepSeek-V4-Pro-0813 has a verified first-party API price: $0.66 per million input tokens and $1.98 per million output tokens. No pay-as-you-go API rate is tracked for Devstral 2 (123B). Devstral 2 (123B) shipped 247 days before DeepSeek-V4-Pro-0813, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Pro-0813 has 1.6T parameters, while Devstral 2 (123B) has 123B. Context windows are 1M (DeepSeek-V4-Pro-0813) vs 256k (Devstral 2 (123B)).
On SWE-Bench Verified, DeepSeek-V4-Pro-0813 leads at 80.6% vs Devstral 2 (123B) at 72.2%.