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 | 120B |
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 | 1M |
| 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. | $0.40Mistral | $0.08DekaLLM |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $2.00Mistral | $0.40DeepInfra |
| $180Mistral | $37.00DeepInfra · bf16 | |
| 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. | 72.2% | 60.5% |
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% | 45.78% |
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% | 31% |
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. | — | 54% |
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. | — | 81.19% |
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better. | — | 31.28% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 79.2% |
AA Intelligence IndexOverall intelligence — Artificial Analysis composite intelligence index across evals | — | 36 |
| Overview | ||
| Company | Mistral | NVIDIA |
| Release date | Dec 9 2025 | Mar 11 2026 |
| Access | Open Weight | Open Source |
| Model details | View model | View model |
Other comparisons
Devstral 2 (123B)vsClaude Haiku 5.5Nemotron 3 SupervsClaude Haiku 5.5Devstral 2 (123B)vsGPT-6.1 SolNemotron 3 SupervsGPT-6.1 SolDevstral 2 (123B)vsGemini 4 ArgonNemotron 3 SupervsGemini 4 ArgonDevstral 2 (123B)vsMuse Spark OpenNemotron 3 SupervsMuse Spark OpenDevstral 2 (123B)vsGrok 4.7Nemotron 3 SupervsGrok 4.7Devstral 2 (123B)vsDeepSeek-V4.1-Flash-1017Nemotron 3 SupervsDeepSeek-V4.1-Flash-1017Frequently asked questions
Devstral 2 (123B) leads Nemotron 3 Super on 3 of the 3 benchmarks they both report (SWE-Bench Verified, SWE-Bench Multilingual, Terminal-Bench 2.0). Devstral 2 (123B) shipped 92 days before Nemotron 3 Super, so benchmark comparisons should account for the intervening progress.
Devstral 2 (123B) has 123B parameters, while Nemotron 3 Super has 120B. Context windows are 256k (Devstral 2 (123B)) vs 1M (Nemotron 3 Super). Devstral 2 (123B) is open weight, while Nemotron 3 Super is open source.
On SWE-Bench Verified, Devstral 2 (123B) leads at 72.2% vs Nemotron 3 Super at 60.5%. On SWE-Bench Multilingual, Devstral 2 (123B) leads at 61.3% vs Nemotron 3 Super at 45.78%. On Terminal-Bench 2.0, Devstral 2 (123B) leads at 32.6% vs Nemotron 3 Super at 31%.