Phi 4
microsoft/phi-4
Phi 4 is a text-only model from Microsoft released on 2024-12-11. It supports a context window of up to 16,384 tokens, has open weights, and does not support tools or reasoning. Pricing is $0.07 per million input tokens and $0.14 per million output tokens.
- structured outputs
- open-weight
At a glance
Specifications
What the API gives you
- Context window
- 16K tokens
- Max output
- 15K tokens
- Input price
- $0.070 / 1M tokens
- Output price
- $0.14 / 1M tokens
- Cache read
- —
- Modalities
- text → text
- Knowledge cutoff
- Jun 30, 2024
- Released
- Dec 11, 2024
Measured evidence
Public benchmark record
- BFCL
- 28.8%2026-07-28
- BFCL
- 28.8%2026-07-27
- BFCL
- 28.8%2026-07-29
- BFCL
- 28.8%2026-07-30
- BFCL
- 28.8%2026-07-31
- BFCL
- 28.8%2026-08-02
- BFCL
- 28.8%2026-08-01
- BFCL
- 28.8%2026-08-05
- BFCL
- 28.8%2026-08-03
- BFCL
- 28.8%2026-08-09
- BFCL
- 28.8%2026-08-04
- BFCL
- 28.8%2026-08-06
- BFCL
- 28.8%2026-08-07
- BFCL
- 28.8%2026-08-10
- BFCL
- 28.8%2026-08-14
- BFCL
- 28.8%2026-08-08
- BFCL
- 28.8%2026-08-11
- BFCL
- 28.8%2026-08-18
- BFCL
- 28.8%2026-08-15
- BFCL
- 28.8%2026-08-12
- BFCL
- 28.8%2026-08-13
- BFCL
- 28.8%2026-08-19
- BFCL
- 28.8%2026-08-16
- BFCL
- 28.8%2026-08-20
- BFCL
- 28.8%2026-08-27
- BFCL
- 28.8%2026-08-17
- BFCL
- 28.8%2026-08-21
- BFCL
- 28.8%2026-08-22
- BFCL
- 28.8%2026-08-23
- BFCL
- 28.8%2026-08-24
- BFCL
- 28.8%2026-08-28
- BFCL
- 28.8%2026-08-25
- BFCL
- 28.8%2026-08-26
- BFCL
- 28.8%2026-08-29
- BFCL
- 28.8%2026-08-30
- BFCL
- 28.8%2026-08-31
- BFCL
- 28.8%2026-09-01
- BFCL
- 28.8%2026-09-02
- BFCL
- 28.8%2026-09-03
- BFCL
- 28.8%2026-09-04
- LMArena Creative Writing
- 1182 Elo2026-07-21
- LMArena Creative Writing
- 1182 Elo2026-07-27
- LMArena Creative Writing
- 1182 Elo2026-07-30
- LMArena Creative Writing
- 1182 Elo2026-08-02
- LMArena Creative Writing
- 1182 Elo2026-08-03
- LMArena Creative Writing
- 1182 Elo2026-08-06
- LMArena Creative Writing
- 1181 Elo2026-08-12
- LMArena Creative Writing
- 1181 Elo2026-08-21
- LMArena Creative Writing
- 1182 Elo2026-08-27
- LMArena Creative Writing
- 1181 Elo2026-09-01
- LMArena Instruction Following
- 1202 Elo2026-07-21
- LMArena Instruction Following
- 1202 Elo2026-07-27
- LMArena Instruction Following
- 1202 Elo2026-07-30
- LMArena Instruction Following
- 1202 Elo2026-08-02
- LMArena Instruction Following
- 1202 Elo2026-08-03
- LMArena Instruction Following
- 1202 Elo2026-08-06
- LMArena Instruction Following
- 1201 Elo2026-08-12
- LMArena Instruction Following
- 1201 Elo2026-08-21
- LMArena Instruction Following
- 1201 Elo2026-08-27
- LMArena Instruction Following
- 1201 Elo2026-09-01
- LMArena Long Query
- 1217 Elo2026-07-21
- LMArena Long Query
- 1217 Elo2026-07-27
- LMArena Long Query
- 1217 Elo2026-07-30
- LMArena Long Query
- 1217 Elo2026-08-02
- LMArena Long Query
- 1217 Elo2026-08-03
- LMArena Long Query
- 1218 Elo2026-08-06
- LMArena Long Query
- 1217 Elo2026-08-12
- LMArena Long Query
- 1217 Elo2026-08-21
- LMArena Long Query
- 1217 Elo2026-08-27
- LMArena Long Query
- 1217 Elo2026-09-01
- LMArena Math
- 1246 Elo2026-07-21
- LMArena Math
- 1246 Elo2026-07-27
- LMArena Math
- 1246 Elo2026-07-30
- LMArena Math
- 1246 Elo2026-08-02
- LMArena Math
- 1246 Elo2026-08-03
- LMArena Math
- 1246 Elo2026-08-06
- LMArena Math
- 1246 Elo2026-08-12
- LMArena Math
- 1246 Elo2026-08-21
- LMArena Math
- 1246 Elo2026-08-27
- LMArena Math
- 1246 Elo2026-09-01
- LMArena Text
- 1256 Elo2026-07-21
- LMArena Text
- 1256 Elo2026-07-27
- LMArena Text
- 1256 Elo2026-07-30
- LMArena Text
- 1256 Elo2026-08-02
- LMArena Text
- 1256 Elo2026-08-10
- LMArena Text
- 1256 Elo2026-08-06
- LMArena Text
- 1256 Elo2026-08-03
- LMArena Text
- 1256 Elo2026-08-21
- LMArena Text
- 1256 Elo2026-08-12
- LMArena Text
- 1256 Elo2026-08-27
- LMArena Text
- 1256 Elo2026-09-01
LMArena data licensed CC-BY-4.0. Aider (Apache-2.0) and SWE-bench (MIT) are open. Design Arena uses blind human preference.
Open-weight model on Hugging Face.
Task evidence
How broad is the support?
Each status is exact to this model version and the selected task. Official release claims remain provisional.
- Best LLM for Business WritingProvisional evidence
70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Marketing CopyProvisional evidence
70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Chat & RoleplayProvisional evidence
70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Roleplay & CharacterProvisional evidence
70% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Tool & Function CallingProvisional evidence
65% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or LMArena Search. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Creative WritingProvisional evidence
60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for WritingProvisional evidence
60% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for a General AssistantProvisional evidence
55% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Logic & Problem SolvingProvisional evidence
50% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Hermes AgentProvisional evidence
44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or SWE-rebench. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for OpenHandsProvisional evidence
44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or SWE-rebench. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for GooseProvisional evidence
44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or SWE-rebench. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for OpenClawProvisional evidence
44% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or SWE-rebench. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Competition MathProvisional evidence
40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Math & ReasoningProvisional evidence
40% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Mathematics or LiveBench Reasoning. 5 of 5 configured evaluation feeds currently have results.
- Best Value LLMProvisional evidence
36% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for AI AgentsProvisional evidence
35% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Agentic Coding or LMArena Agent. 7 of 7 configured evaluation feeds currently have results.
- Best LLM for JSON & Schema OutputProvisional evidence
33% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Structured Data ExtractionProvisional evidence
33% of intended weighted task evidence across 2 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Structured-output eval. 4 of 5 configured evaluation feeds currently have results.
- Best Open-Weight LLM for Self-HostingProvisional evidence
33% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Fast SummariesProvisional evidence
29% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Literary TranslationProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Best Cheap LLM for High VolumeProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Everyday TranslationProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Cheapest Capable LLMProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or Route reliability. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for TranslationProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LiveBench Instruction Following or LiveBench Language. 4 of 5 configured evaluation feeds currently have results.
- Best LLM for Massive ContextProvisional evidence
25% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Document or LongBench v2. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Document SummarizationProvisional evidence
20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for SummarizationProvisional evidence
20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LiveBench Instruction Following. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for Computer & Browser UseProvisional evidence
20% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from LMArena Agent or OSWorld Verified. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for RAG & Long ContextProvisional evidence
18% of intended weighted task evidence across 1 independent source; broader independent confirmation is still missing from FACTS Grounding or LMArena Document. 4 of 6 configured evaluation feeds currently have results.
- Best LLM for High-Volume ExtractionProvisional evidence
13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Data Analysis or Route reliability. 2 of 4 configured evaluation feeds currently have results.
- Best LLM for Code CompletionProvisional evidence
13% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 4 of 4 configured evaluation feeds currently have results.
- Best LLM for Debugging & Code ReviewProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 6 of 6 configured evaluation feeds currently have results.
- Best LLM for CodingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or Design Arena Coding. 8 of 8 configured evaluation feeds currently have results.
- Best LLM for OCR & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Image UnderstandingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LiveBench Reasoning or LMArena Vision. 3 of 4 configured evaluation feeds currently have results.
- Best LLM for Vision & DocumentsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Grounded RAGProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from CRAG or FACTS Grounding. 2 of 5 configured evaluation feeds currently have results.
- Best LLM for Charts & DiagramsProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best LLM for Document ParsingProvisional evidence
10% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from LMArena Document or LMArena Vision. 3 of 5 configured evaluation feeds currently have results.
- Best Local LLM for CodingProvisional evidence
6% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Aider Polyglot or LiveBench Coding. 5 of 5 configured evaluation feeds currently have results.
- Best LLM for Agentic CodingProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Design Arena Full-stack or Design Arena Web-apps. 8 of 8 configured evaluation feeds currently have results.
- Best LLM for Autonomous AgentsProvisional evidence
5% of intended weighted task evidence across 0 independent sources; broader independent confirmation is still missing from Frontier-Bench or LiveBench Agentic Coding. 7 of 7 configured evaluation feeds currently have results.
Recommendation coverage
Where this model appears
Not currently included in a published recommendation. The task evidence panel above says exactly what is confirmed and what is still missing.
Catalog data via OpenRouter and models.dev. Last checked within the past 6 hours.