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. | 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. | 256k | 400k |
| API pricingUSD per 1M tokens · lower wins | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | — | $12.50 |
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. | — | $75.00 |
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. | — | $1.25 |
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. | $0.40Mistral | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.00Mistral | — |
| $180Mistral | $6,250OpenAI list price | |
| These models have no shared benchmark scores. | ||
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. | 72.2% | — |
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 2.0Agentic 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% | — |
Advanced Cybersecurity Completion Rate (Internal)Agentic tasks — share of advanced cyber requests the model responds to; measures willingness to answer, not exploit success | — | 95% |
| Overview | ||
| Company | Mistral | OpenAI |
| Release date | Dec 9 2025 | Aug 10 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Devstral 2 (123B)vsClaude Fable 5.2GPT-5.6 CybervsClaude Fable 5.2Devstral 2 (123B)vsGemini 4 CarbonGPT-5.6 CybervsGemini 4 CarbonDevstral 2 (123B)vsMuse EmberGPT-5.6 CybervsMuse EmberDevstral 2 (123B)vsGrok 4.7GPT-5.6 CybervsGrok 4.7Devstral 2 (123B)vsDeepSeek-V4.1-Flash-1017GPT-5.6 CybervsDeepSeek-V4.1-Flash-1017Devstral 2 (123B)vsKimi K3.1 ProGPT-5.6 CybervsKimi K3.1 ProFrequently asked questions
Devstral 2 (123B) and GPT-5.6 Cyber don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Only GPT-5.6 Cyber has a verified first-party price for an available API: $12.50 per million input tokens and $75.00 per million output tokens. No currently available pay-as-you-go API rate is tracked for Devstral 2 (123B). Devstral 2 (123B) shipped 244 days before GPT-5.6 Cyber, so benchmark comparisons should account for the intervening progress.
Context windows are 256k (Devstral 2 (123B)) vs 400k (GPT-5.6 Cyber). Devstral 2 (123B) is open weight, while GPT-5.6 Cyber is closed.
Direct benchmark comparisons are unavailable — Devstral 2 (123B) and GPT-5.6 Cyber don't publish scores on any of the same benchmarks.