phi-4 vs Mistral Large 3
At a Glance
| Compare |  phi-4Microsoft |  Mistral Large 3Mistral AI |
|---|---|---|
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.070Deepinfra ↗ · Sep 22, 2026 | $0.25Openrouter ↗ · Sep 22, 2026 |
| Output priceFrom · USD / 1M tokens | $0.14Deepinfra ↗ · Sep 22, 2026 | $0.75Openrouter ↗ · Sep 22, 2026 |
| Context windowMaximum documented tokens | 16K | 262K |
| Model facts checked | Aug 28, 2026View model evidence → | Aug 29, 2026View model evidence → |
Token prices are the lowest available sourced USD rates; input and output may use different providers. Cost ranking estimates output spend on LiveBench, not a full request bill. Ranking methodology →
Available Benchmarks
| Benchmark | phi-4 | Mistral Large 3 |
|---|---|---|
| LMArena Text Arenatext-2026-09-13-d25aabda0010 · arena_rating · leader | 1,216.6085% of row best · rating · phi-4; 95% CI [1212.02298793, 1221.18318193]; votes 24126; rank 300 | 1,427.04100% of row best · rating · mistral-large-3; 95% CI [1424.05319044, 1430.03486299]; votes 69028; rank 91 |
| Overall ResultCounted from the protocol-matched rows above | 0 benchmark winsNo overall winner | 1 benchmark winNo overall winner |
Third-party benchmark Only like-for-like primary-publisher results are shown; raw scores, relative scores, configuration, and token spend remain visible.
Side-by-Side Facts
| Field | phi-4 | Mistral Large 3 |
|---|---|---|
| Developer | Microsoft | Mistral AI |
| Family | Phi 4 | Mistral Large 3 |
| Model | phi-4 | Mistral Large 3 |
| Version | phi-4 | Mistral Large 3 |
| Lifecycle | active | active |
| Released | 2024-12-12 | 2025-12-02 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text | Text, Image, Document |
| Output modalities | Text | Text |
| Context window | 16K | 262K |
| Total parameters | 14.7B | 675B |
| Active parameters | Unknown | 41B |
| License | mit | Apache-2.0 |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Hugging Face (Standard), Openrouter (Standard) | Mistral AI (Standard), Openrouter (Standard) |
| Capabilities | chat, generation | agents, chat, generation, structured_outputs, tools, vision |
phi-4 Capabilities
Mistral Large 3 Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
phi-4Microsoft | vs |  GPT-6 AstraOpenAI | cross-developer peerstext |
Claude Fable 5.1Anthropic | vs | phi-4Microsoft | cross-developer peerstext |
 Gemini 3.1 ProGoogle DeepMind | vs | phi-4Microsoft | cross-developer peerstext |
 DeepSeek-V4.1-FlashDeepSeek | vs | phi-4Microsoft | cross-developer peerstext |
phi-4Microsoft | vs | Grok 4.6xAI | cross-developer peerstext |
phi-4Microsoft | vs |  Qwen3.8-MaxQwen | cross-developer peerstext |
phi-4Microsoft | vs |  Kimi-K3Moonshot AI | cross-developer peerstext |
 MiniMax-M3MiniMax | vs | phi-4Microsoft | cross-developer peerstext |
phi-4Microsoft | vs |  GLM-5.3Z.ai | cross-developer peerstext |
phi-4Microsoft | vs |  Hy4 previewTencent | cross-developer peerstext |
Seed 2.1 ProByteDance Seed | vs | phi-4Microsoft | cross-developer peerstext |
| vs | phi-4Microsoft | cross-developer peerstext |
Primary Evidence
Sources and Freshness
Questions
phi-4 vs Mistral Large 3 FAQs
Is phi-4 or Mistral Large 3 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both phi-4 and Mistral Large 3, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, phi-4 or Mistral Large 3?+
phi-4 is $0.070 and Mistral Large 3 is $0.25 per million tokens, so phi-4 is cheaper on this metric. phi-4 is $0.14 and Mistral Large 3 is $0.75 per million tokens, so phi-4 is cheaper on this metric.
Which has a larger context window, phi-4 or Mistral Large 3?+
Mistral Large 3 has the larger sourced context window. phi-4 supports 16K and Mistral Large 3 supports 262K.
Which performs better in benchmarks, phi-4 or Mistral Large 3?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can phi-4 or Mistral Large 3 be self-hosted?+
Both models have the same recorded self-hosting status: supported. phi-4 is open weight; Mistral Large 3 is open weight.
Can phi-4 and Mistral Large 3 understand images?+
phi-4 is not documented with image input; Mistral Large 3 is documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, phi-4 or Mistral Large 3?+
Neither has a larger sourced maximum output. phi-4 is — and Mistral Large 3 is —.
Do phi-4 and Mistral Large 3 support reasoning and tool use?+
phi-4: none of these features are definitively sourced. Mistral Large 3: tool calling and image input. Feature support does not establish relative quality.
Which is available from more inference providers, phi-4 or Mistral Large 3?+
phi-4 has 3 sourced provider routes; Mistral Large 3 has 2, so phi-4 has broader tracked availability.
Which offers better value, phi-4 or Mistral Large 3?+
There is no universal value winner. Compare the input and output prices above with the matched benchmark result for your workload: cheaper tokens can be offset by different quality, token usage, latency, or provider availability.