Mistral Large 2 / Mixtral family vs xAI Grok 4
Mistral Large 2 / Mixtral family vs xAI Grok 4: side-by-side comparison of two foundation models (substrate reference) systems — architecture, taxonomy, license, pricing, MCP/A2A support, and direct edges.
Cost & capability
| Mistral Large 2 / Mixtral family | xAI Grok 4 | |
|---|---|---|
| Capability band | competent | frontier |
| Capability composite | 74 | 88 |
| Cost tier | mid | premium |
| $/Mtok input | 2 | 3 |
| $/Mtok output | 6 | 15 |
| Use cases | Scoped Agentic, Code Generation Focused, Analytical Summarization, Memory Augmented Chat | Scoped Agentic, Code Generation Focused, Analytical Summarization, Memory Augmented Chat |
Where they differ (16)
Rows where both sides have data and the values disagree — the shortlist of dimensions that actually distinguish these two systems.
| Mistral Large 2 / Mixtral family | xAI Grok 4 | |
|---|---|---|
| Capability band | competent | frontier |
| Capability composite | 74 | 88 |
| Cost tier | mid | premium |
| $/Mtok input | 2 | 3 |
| $/Mtok output | 6 | 15 |
| Type | Frontier model family — Mistral Large 2 + Mixtral 8x22B + Mistral Small 3 | Frontier foundation model from xAI (Grok 4) |
| Created | 2023-04 (Mistral AI founded); 2023-09 (Mistral 7B); 2023-12 (Mixtral 8x7B); 2024-04 (Mixtral 8x22B); 2024-07 (Large 2); 2025-01 (Small 3) | 2023-07 (xAI founded); 2023-11 (Grok 1); 2024-08 (Grok 2); 2025-02 (Grok 3); 2025-07 (Grok 4) |
| Latest release | Mistral Medium 3.5 (2026-04-29); Mistral Large 4 in public preview (2026-10), weights not yet released | Grok 4.7 (2026-09-21) |
| License | Mixed: Mistral Research License MRL-0.1, non-OSI (Mistral Large 2 / 2407 and 2411 weights; commercial use needs a separate licence); Mistral Non-Production License (Codestral 22B); Modified MIT (Medium 3.5 weights); Apache-2.0 (Mixtral 8x22B, Small / Large 3 / other open models); some models API-only proprietary | proprietary (Grok 4); older open weights: Grok-1 (Apache-2.0) and Grok 2 (Grok 2 Community License) |
| GitHub | github.com/mistralai/mistral-inference + various model repos; cumulative ~30k+ stars | github.com/xai-org/grok-1 ~50k stars (older Grok 1 weights only) |
| Pricing | API: Mistral Medium 3.5 $1.50 in / $7.50 out per 1M; Mistral Large 3 $0.50 / $1.50; batch 50% off; open weights free | API: Grok 4.7 $2 in / $6 out per 1M (prompts ‹200k tokens; $4 / $12 above); consumer via X / SuperGrok subscriptions |
| Funding | €2.4B+ total raised (~$2.7B USD); €11.7B valuation Series C Sept-2025 (ASML lead); a16z, General Catalyst, DST, Lightspeed, Nvidia investors | xAI merged into SpaceX (announced 2026-02-02, combined ~$1T); prior $20B Series E (2026-01, ~$230B valuation) |
| Deployment | Both (self-hosted + managed cloud) — La Plateforme / Mistral Studio API, open weights for self-hosting, plus Azure / AWS / GCP marketplaces | Managed-only — xAI API, grok.com, X apps; also via Amazon Bedrock, Microsoft Foundry, GitHub Copilot (older Grok-1 / Grok 2 weights downloadable) |
| API surface | REST + SDK (Python, JS); HuggingFace Transformers for open-weights; Azure AI Foundry; AWS Bedrock; GCP Vertex | REST + SDK (Python, TS); OpenAI-compatible |
| Optimised for | EU-hosted enterprise; sovereign AI for European governments; cost-effective hybrid (open + commercial) family | real-time information (X firehose access); contrarian / less-filtered persona; benchmark frontier-tier reasoning |
| Anti-fit | not for users needing absolute frontier capability (lags Claude/GPT/Gemini on most benchmarks); not for non-Apache OSS users who want weight openness on Large 2 | X/Twitter integration is unique selling point but creates Musk-political-association risk; weights closed at Grok 4 tier; no on-prem; less polished tool-use than Claude/GPT |
At a glance
| Mistral Large 2 / Mixtral family | xAI Grok 4 | |
|---|---|---|
| Section | Foundation models (substrate reference) | Foundation models (substrate reference) |
| Tier | T1 | T1 |
| Type | Frontier model family — Mistral Large 2 + Mixtral 8x22B + Mistral Small 3 | Frontier foundation model from xAI (Grok 4) |
| Created | 2023-04 (Mistral AI founded); 2023-09 (Mistral 7B); 2023-12 (Mixtral 8x7B); 2024-04 (Mixtral 8x22B); 2024-07 (Large 2); 2025-01 (Small 3) | 2023-07 (xAI founded); 2023-11 (Grok 1); 2024-08 (Grok 2); 2025-02 (Grok 3); 2025-07 (Grok 4) |
| Latest release | Mistral Medium 3.5 (2026-04-29); Mistral Large 4 in public preview (2026-10), weights not yet released | Grok 4.7 (2026-09-21) |
| License | Mixed: Mistral Research License MRL-0.1, non-OSI (Mistral Large 2 / 2407 and 2411 weights; commercial use needs a separate licence); Mistral Non-Production License (Codestral 22B); Modified MIT (Medium 3.5 weights); Apache-2.0 (Mixtral 8x22B, Small / Large 3 / other open models); some models API-only proprietary | proprietary (Grok 4); older open weights: Grok-1 (Apache-2.0) and Grok 2 (Grok 2 Community License) |
| GitHub | github.com/mistralai/mistral-inference + various model repos; cumulative ~30k+ stars | github.com/xai-org/grok-1 ~50k stars (older Grok 1 weights only) |
| Pricing | API: Mistral Medium 3.5 $1.50 in / $7.50 out per 1M; Mistral Large 3 $0.50 / $1.50; batch 50% off; open weights free | API: Grok 4.7 $2 in / $6 out per 1M (prompts ‹200k tokens; $4 / $12 above); consumer via X / SuperGrok subscriptions |
| Funding | €2.4B+ total raised (~$2.7B USD); €11.7B valuation Series C Sept-2025 (ASML lead); a16z, General Catalyst, DST, Lightspeed, Nvidia investors | xAI merged into SpaceX (announced 2026-02-02, combined ~$1T); prior $20B Series E (2026-01, ~$230B valuation) |
| Deployment | Both (self-hosted + managed cloud) — La Plateforme / Mistral Studio API, open weights for self-hosting, plus Azure / AWS / GCP marketplaces | Managed-only — xAI API, grok.com, X apps; also via Amazon Bedrock, Microsoft Foundry, GitHub Copilot (older Grok-1 / Grok 2 weights downloadable) |
| API surface | REST + SDK (Python, JS); HuggingFace Transformers for open-weights; Azure AI Foundry; AWS Bedrock; GCP Vertex | REST + SDK (Python, TS); OpenAI-compatible |
| Optimised for | EU-hosted enterprise; sovereign AI for European governments; cost-effective hybrid (open + commercial) family | real-time information (X firehose access); contrarian / less-filtered persona; benchmark frontier-tier reasoning |
| Anti-fit | not for users needing absolute frontier capability (lags Claude/GPT/Gemini on most benchmarks); not for non-Apache OSS users who want weight openness on Large 2 | X/Twitter integration is unique selling point but creates Musk-political-association risk; weights closed at Grok 4 tier; no on-prem; less polished tool-use than Claude/GPT |
Pros & cons
Mistral Large 2 / Mixtral family
Pros: Strongest EU AI-sovereignty story (Paris-HQ, GDPR-native, French gov customer); Apache 2.0 open weights for Mixtral / Small / Codestral; €2.7B raised at €11.7B valuation.
Cons: Trails US/China frontier labs on raw benchmarks; Large 2 not open-weights; smaller training-compute budget; multimodal less mature than peers.
xAI Grok 4
Pros: Real-time X/Twitter data access (unique among frontier models); Colossus supercluster gives independent training capacity; ARC-AGI-2 leader (vendor-stated).
Cons: Owner Musk political controversy creates enterprise procurement headwind; closed weights at Grok 4; no on-prem; less mature ecosystem than Anthropic/OpenAI/Google.