LLM Pricing
← All models

Phi-4-reasoning

azure_ai · chat model

Phi-4-reasoning is listed here as a chat model from azure_ai. This page shows simple API pricing, token limits, and capability flags so you can compare it with similar options.

Input
$0.1250 / 1M tokens
Output
$0.5000 / 1M tokens
Cached input
N/A
Context
4.1K
Compare similar models Open a side-by-side table using this route and the closest match from the current catalog. Estimate workload cost Use token counts, request volume, and cache assumptions to check a realistic monthly bill. Explore benchmark family Review source-backed benchmark rows for the attached base model family. Check data sources See how pricing data is generated and what still needs provider verification.

Quick read

Best for

Use this page when you need a fast view of cost, context size, and supported features before testing the model in your own workload.

Things to verify

Always check the provider page for discounts, cache pricing, region rules, and any model limits that may not appear in public metadata.

Pricing

Item Price
Input
$0.1250 / 1M tokens
Output
$0.5000 / 1M tokens
Embedding
$0.1250 / 1M tokens

Token limits

Context window
4.1K
Max input tokens
32.8K
Max output tokens
4.1K
Max tokens
4.1K

Capabilities

Capability Supported
Vision —
Function calling ✅
Parallel function calling —
Tool choice ✅
Prompt caching —
Reasoning ✅
Response schema —
System messages —
Audio input —
Audio output —
Web search —
PDF input —
Video input —

Benchmarks

Most benchmark rows are attached to the base model family rather than this provider route. Open benchmark explorer

Benchmark Score Metric Scope Checked Source
MMLU 84.8 Popular Aggregated Benchmark Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
GPQA 56.1 Science Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
HumanEval 82.6 Code Generation Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
DROP 75.5 Reasoning Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
MMLU 84.8 Popular Aggregated Benchmark Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
GPQA 56.1 Science Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
HumanEval 82.6 Code Generation Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
SimpleQA 3.0 Factual Knowledge Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link
DROP 75.5 Reasoning Base model: Phi-4 (phi-4 (14B)) 2026-05-31 Link

Sources

Source links
Pricing data LiteLLM model cost map
Synced at 2026-05-28

Docs

Official docs

Similar models

This list is ranked by overall similarity. Use filters to emphasize the lens that matters most for the replacement you are making.

Comparing from
Model Cost Input shape Features Context Why it is close
Phi-4-reasoning
azure_ai
In $0.1250 / 1M tokens
Out $0.5000 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
4.1K
Current model
Reference row

Overall blends cost, modality overlap, capabilities, and context.

Model Cost Input shape Features Context Why it is close
command-r
vercel_ai_gateway
In $0.1500 / 1M tokens
Out $0.6000 / 1M tokens
text
Output: text
Function callingTool choice
4.1K
Text covered
Overall 88%
Phi-3-mini-128k-instruct
azure_ai
In $0.1300 / 1M tokens
Out $0.5200 / 1M tokens
text
Output: text
Tool choice
4.1K
Same provider
Overall 83%
Phi-3-mini-4k-instruct
azure_ai
In $0.1300 / 1M tokens
Out $0.5200 / 1M tokens
text
Output: text
Tool choice
4.1K
Same provider
Overall 83%
Phi-3.5-mini-instruct
azure_ai
In $0.1300 / 1M tokens
Out $0.5200 / 1M tokens
text
Output: text
Tool choice
4.1K
Same provider
Overall 83%
qwen3-235b-a22b-thinking-2507
OpenRouter
In $0.1100 / 1M tokens
Out $0.6000 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
262.1K
Text covered
Overall 82%
llama-2-13b
replicate
In $0.1000 / 1M tokens
Out $0.5000 / 1M tokens
text
Output: text
Tool choice
4.1K
Text covered
Overall 82%
Phi-4
azure_ai
In $0.1250 / 1M tokens
Out $0.5000 / 1M tokens
text
Output: text
Function callingTool choice
16.4K
Same provider
Overall 80%
Phi-3-small-128k-instruct
azure_ai
In $0.1500 / 1M tokens
Out $0.6000 / 1M tokens
text
Output: text
Tool choice
4.1K
Same provider
Overall 80%
Phi-3-small-8k-instruct
azure_ai
In $0.1500 / 1M tokens
Out $0.6000 / 1M tokens
text
Output: text
Tool choice
4.1K
Same provider
Overall 80%
qwen.qwen3-235b-a22b-2507-v1:0
bedrock_converse
In $0.2200 / 1M tokens
Out $0.8800 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
131.1K
Text covered
Overall 77%
DeepSeek-V3.1
deepinfra
In $0.2700 / 1M tokens
Out $1.0000 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 75%
qwen.qwen3-coder-480b-a35b-v1:0
bedrock_converse
In $0.2200 / 1M tokens
Out $1.8000 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
65.5K
Text covered
Overall 74%
deepseek.v3-v1:0
bedrock_converse
In $0.5800 / 1M tokens
Out $1.6800 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
81.9K
Text covered
Overall 71%
deepseek.v3.2
Bedrock
In $0.6200 / 1M tokens
Out $1.8500 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 70%
deepseek.v3.2
Bedrock
In $0.7400 / 1M tokens
Out $2.2200 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 69%
deepseek.v3.2
Bedrock
In $0.7400 / 1M tokens
Out $2.2200 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 69%
deepseek.v3.2
Bedrock
In $0.7400 / 1M tokens
Out $2.2200 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 69%
deepseek.v3.2
Bedrock
In $0.7400 / 1M tokens
Out $2.2200 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 69%
deepseek.v3.2
Bedrock
In $0.7400 / 1M tokens
Out $2.2200 / 1M tokens
text
Output: text
Function callingTool choiceReasoning
163.8K
Text covered
Overall 69%
Phi-3.5-vision-instruct
azure_ai
In $0.1300 / 1M tokens
Out $0.5200 / 1M tokens
text image
Output: text
Tool choice
4.1K
Same provider
Overall 68%
gpt-oss-120b
baseten
In $0.1000 / 1M tokens
Out $0.5000 / 1M tokens
text
Output: unknown
Low overlap
N/A
Partial I/O overlap
Overall 18%
Missing text