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. | 24B | 49B |
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 | 128k |
| 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. | 68% | — |
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. | 55.7% | — |
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. | — | 73.58% |
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. | 22.5% | — |
Humanity's Last Exam · text only, no toolsMultidisciplinary reasoning — accuracy on the 2,158 text-only questions without tools | — | 7.64% |
MATH-500Reasoning — accuracy on the 500-problem MATH subset | — | 97.4% |
AIME 2024Reasoning — single-answer accuracy on the 2024 invitational mathematics exam without tools | — | 87.5% |
AIME 2025Reasoning — single-answer accuracy on the 2025 invitational mathematics exam without tools | — | 82.71% |
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | — | 71.97% |
MMLU-ProKnowledge — accuracy on the ten-option MMLU-Pro benchmark | — | 79.53% |
| Overview | ||
| Company | Mistral | NVIDIA |
| Release date | Dec 9 2025 | Jul 25 2025 |
| Access | Open Weight | Open Weight |
| Model details | View model | View model |
Other comparisons
Devstral Small 2 (24B)vsClaude Fable 5.2Llama Nemotron Super 49B v1.5vsClaude Fable 5.2Devstral Small 2 (24B)vsGPT-6.2 AstraLlama Nemotron Super 49B v1.5vsGPT-6.2 AstraDevstral Small 2 (24B)vsGemini 4 NeonLlama Nemotron Super 49B v1.5vsGemini 4 NeonDevstral Small 2 (24B)vsMuse EmberLlama Nemotron Super 49B v1.5vsMuse EmberDevstral Small 2 (24B)vsGrok 4.7Llama Nemotron Super 49B v1.5vsGrok 4.7Devstral Small 2 (24B)vsDeepSeek-V4.1-Flash-1017Llama Nemotron Super 49B v1.5vsDeepSeek-V4.1-Flash-1017Frequently asked questions
Devstral Small 2 (24B) and Llama Nemotron Super 49B v1.5 don't publish scores on any of the same benchmarks, so there's no direct head-to-head comparison. Llama Nemotron Super 49B v1.5 shipped 137 days before Devstral Small 2 (24B), so benchmark comparisons should account for the intervening progress.
Devstral Small 2 (24B) has 24B parameters, while Llama Nemotron Super 49B v1.5 has 49B. Context windows are 256k (Devstral Small 2 (24B)) vs 128k (Llama Nemotron Super 49B v1.5).
Direct benchmark comparisons are unavailable — Devstral Small 2 (24B) and Llama Nemotron Super 49B v1.5 don't publish scores on any of the same benchmarks.