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. | 128B | — |
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 | — |
| 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. | $1.50 | $1.25 |
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. | $7.50 | $2.50 |
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. | $0.15 | $0.20 |
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. | $1.50Mistral | — |
Cheapest outputLowest output rate across every provider on OpenRouter, the lab included. May come from a different provider than the cheapest input. | $7.50Mistral | — |
| $675Mistral list price | $375SpaceXSI list price | |
| 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. | 77.6% | 76.7% |
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. | — | 56% |
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. | — | 53.3% |
| Overview | ||
| Company | Mistral | SpaceXSI |
| Release date | Apr 29 2026 | Feb 17 2026 |
| Access | Open Weight | Closed |
| Model details | View model | View model |
Other comparisons
Mistral Medium 3.5vsClaude Haiku 5.5Grok 4.20 BetavsClaude Haiku 5.5Mistral Medium 3.5vsGPT-6.1 TerraGrok 4.20 BetavsGPT-6.1 TerraMistral Medium 3.5vsGemini 4 CarbonGrok 4.20 BetavsGemini 4 CarbonMistral Medium 3.5vsMuse Spark OpenGrok 4.20 BetavsMuse Spark OpenMistral Medium 3.5vsDeepSeek-V4.1-Flash-1017Grok 4.20 BetavsDeepSeek-V4.1-Flash-1017Mistral Medium 3.5vsKimi K3.1 ProGrok 4.20 BetavsKimi K3.1 ProFrequently asked questions
Mistral Medium 3.5 leads Grok 4.20 Beta on 1 of the 1 benchmark they both report (SWE-Bench Verified). Grok 4.20 Beta is cheaper on both input and output: $1.25 vs $1.50 per million input tokens, and $2.50 vs $7.50 per million output tokens. Figures are base-tier rates. Grok 4.20 Beta shipped 71 days before Mistral Medium 3.5, so benchmark comparisons should account for the intervening progress.
Mistral Medium 3.5 is open weight, while Grok 4.20 Beta is closed.
On SWE-Bench Verified, Mistral Medium 3.5 leads at 77.6% vs Grok 4.20 Beta at 76.7%.