MiniMax M3 vs phi-4
At a Glance
| Compare |  MiniMax M3MiniMax |  phi-4Microsoft |
|---|---|---|
| Intelligence, Cost, and Efficiency | ||
| IntelligenceHigher is better · MM Intelligence v2.5 | #35 of 4632.6 score · 2/3 sources · provisional · missing ARC-AGI-2 · full-core range 21.8–55.1 | UnrankedNot in the 46-model eligible cohort |
| CostLower is better · Published-token output estimate | #3 of 44$0.017 per LiveBench case | UnrankedNot in the 44-model eligible cohort |
| EfficiencyHigher is better · MM Efficiency v1.5 | #10 of 3859.2 score · 2/3 sources · provisional · missing ARC-AGI-2 · full-core range 53.8–70.4 | UnrankedNot in the 38-model eligible cohort |
| Pricing and Limits | ||
| Input priceFrom · USD / 1M tokens | $0.28Deepinfra ↗ · Sep 22, 2026 | $0.070Deepinfra ↗ · Sep 22, 2026 |
| Output priceFrom · USD / 1M tokens | $1.10Deepinfra ↗ · Sep 22, 2026 | $0.14Deepinfra ↗ · Sep 22, 2026 |
| Context windowMaximum documented tokens | 1,049K | 16K |
| Model facts checked | Aug 28, 2026View model evidence → | Aug 28, 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 | MiniMax-M3 | phi-4 |
|---|---|---|
| LMArena Text Arenatext-2026-09-13-d25aabda0010 · arena_rating · leader | 1,433.50100% of row best · rating · minimax-m3; 95% CI [1429.31077607, 1437.68800840]; votes 48540; rank 82 | 1,216.6085% of row best · rating · phi-4; 95% CI [1212.02298793, 1221.18318193]; votes 24126; rank 300 |
| Overall ResultCounted from the protocol-matched rows above | 1 benchmark winNo overall winner | 0 benchmark winsNo 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 | MiniMax-M3 | phi-4 |
|---|---|---|
| Developer | MiniMax | Microsoft |
| Family | Minimax M3 | Phi 4 |
| Model | MiniMax-M3 | phi-4 |
| Version | MiniMax-M3 | phi-4 |
| Lifecycle | active | active |
| Released | 2026-06-01 | 2024-12-12 |
| Knowledge cutoff | Unknown | Unknown |
| Input modalities | Text, Image | Text |
| Output modalities | Text | Text |
| Context window | 1,049K | 16K |
| Total parameters | 427B | 14.7B |
| Active parameters | 23B | Unknown |
| License | other | mit |
| Open weights | Yes | Yes |
| API available | Yes | Yes |
| Self-hostable | Yes | Yes |
| Provider access | Deepinfra (Standard), Fireworks Ai (Standard), Hugging Face (Standard), Openrouter (Standard), Together Ai (Standard) | Deepinfra (Standard), Hugging Face (Standard), Openrouter (Standard) |
| Capabilities | chat, generation, reasoning, tools | chat, generation |
MiniMax M3 Capabilities
phi-4 Capabilities
Related Comparisons
All text comparisons →| A | Pair | B | Context |
|---|---|---|---|
MiniMax-M2.7MiniMax | vs | MiniMax-M3MiniMax | family variantstext |
MiniMax-M3MiniMax | vs |  GPT-6 AstraOpenAI | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs | Claude Fable 5.1Anthropic | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs |  Gemini 3.1 ProGoogle DeepMind | cross-developer peerstext |
MiniMax-M3MiniMax | vs |  DeepSeek-V4.1-FlashDeepSeek | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs | Grok 4.6xAI | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs |  Qwen3.8-MaxQwen | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs |  Kimi-K3Moonshot AI | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs |  GLM-5.3Z.ai | cross-developer peerstext |
MiniMax-M3MiniMax | vs |  Hy4 previewTencent | cross-developer peerstext |
MiniMax-M3MiniMax | vs | Seed 2.1 ProByteDance Seed | cross-developer peersimage, text |
MiniMax-M3MiniMax | vs |  Mistral Large 3Mistral AI | cross-developer peersimage, text |
Primary Evidence
Sources and Freshness
Questions
MiniMax M3 vs phi-4 FAQs
Is MiniMax M3 or phi-4 better for coding?+
This comparison does not currently contain a protocol-matched coding benchmark for both MiniMax M3 and phi-4, so Model Markets cannot name a coding leader from pricing, context size, or capability labels alone.
Which is cheaper, MiniMax M3 or phi-4?+
MiniMax M3 is $0.28 and phi-4 is $0.070 per million tokens, so phi-4 is cheaper on this metric. MiniMax M3 is $1.10 and phi-4 is $0.14 per million tokens, so phi-4 is cheaper on this metric.
Which has a larger context window, MiniMax M3 or phi-4?+
MiniMax M3 has the larger sourced context window. MiniMax M3 supports 1,049K and phi-4 supports 16K.
Which performs better in benchmarks, MiniMax M3 or phi-4?+
There is no overall benchmark winner: At least two independently verified, protocol-matched benchmarks are required for an overall winner.
Can MiniMax M3 or phi-4 be self-hosted?+
Both models have the same recorded self-hosting status: supported. MiniMax M3 is open weight; phi-4 is open weight.
Can MiniMax M3 and phi-4 understand images?+
MiniMax M3 is documented with image input; phi-4 is not documented with image input. This reflects supported input modalities, not vision quality.
Which can generate longer answers, MiniMax M3 or phi-4?+
Neither has a larger sourced maximum output. MiniMax M3 is — and phi-4 is —.
Do MiniMax M3 and phi-4 support reasoning and tool use?+
MiniMax M3: reasoning, tool calling, and image input. phi-4: none of these features are definitively sourced. Feature support does not establish relative quality.
Which is available from more inference providers, MiniMax M3 or phi-4?+
MiniMax M3 has 5 sourced provider routes; phi-4 has 3, so MiniMax M3 has broader tracked availability.
Which offers better value, MiniMax M3 or phi-4?+
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.