LLM Model LandscapeFoundational

Multimodal Input Processing

The GPT-6 tiers, what reasoning effort replaced, when to use the Responses API over Chat Completions, how the bill is calculated, and what is different on Azure.

By Amit UpadhyayPublished 13 Oct 202611 min

Step 1 / 4

Speed

What's happening

GPT-6.1 Sol Model

gpt-6.1-sol: the default GPT-6 model — text and image in, text out

💡 Analogy

Like a Swiss Army knife vs individual tools — one model handles text and images natively.

GPT-6.1 Sol is the balanced tier of the GPT-6 family: near top-tier quality at $2 input / $10 output per 1M tokens. It takes text and images and always reasons before answering.

Console Output

> model: "gpt-6.1-sol"
> Context: 1.05M tokens
> max_completion_tokens: 2000 (output + reasoning cap)
> Response mode: message content

●gpt-6.1-sol.py

GPT-6.1 Sol ModelL6–9

1from openai import OpenAI

2import base64

4client = OpenAI()

6# Text: multimodal by default, text only here

7response = client.chat.completions.create(

8 model="gpt-6.1-sol",

9 messages=[

10 {"role": "system", "content": "You are a helpful assistant."},

11 {"role": "user", "content": "Explain attention in transformers in 2 sentences."},

12 ],

13 max_completion_tokens=2000, # includes reasoning tokens

14 reasoning_effort="low", # low | medium | high | xhigh | max

15)

16print(response.choices[0].message.content)

17print(f"Tokens: {response.usage.total_tokens}")

19# Vision: pass image URL or base64

20with open("diagram.png", "rb") as f:

21 image_b64 = base64.b64encode(f.read()).decode()

23vision_response = client.chat.completions.create(

24 model="gpt-6.1-sol",

25 messages=[{

26 "role": "user",

27 "content": [

28 {"type": "text", "text": "What does this architecture diagram show?"},

29 {"type": "image_url", "image_url": {

30 "url": f"data:image/png;base64,{image_b64}",

31 "detail": "high", # high = 4096 tokens for detail

32 }},

33 ],

34 }],

35)

36print(vision_response.choices[0].message.content)

36 lines

Steps · 4

What's happening

GPT-6.1 Sol Model

gpt-6.1-sol: the default GPT-6 model — text and image in, text out

💡 Analogy

Like a Swiss Army knife vs individual tools — one model handles text and images natively.

GPT-6.1 Sol is the balanced tier of the GPT-6 family: near top-tier quality at $2 input / $10 output per 1M tokens. It takes text and images and always reasons before answering.

Console Output

> model: "gpt-6.1-sol"
> Context: 1.05M tokens
> max_completion_tokens: 2000 (output + reasoning cap)
> Response mode: message content

Next: Test →

GPT-6 OpenAI reasoning-effort Responses-API model-selection

← PreviousClaude: Anthropic's Model Family in Practice

Continue reading

All LLM Model Landscape articles →

LLM Model Landscape11 min

Claude: Anthropic's Model Family in Practice

The four Claude tiers, adaptive thinking and effort, the tool-use rules that differ from other providers, prompt caching, and where you can run it.

LLM Model Landscape11 min

Llama: Open-Weight Models for Production

The Llama lineup and licence, serving with Ollama and vLLM, GPU memory sizing, and an honest look at when a hosted API is still the better choice.

LLM Model Landscape10 min

Evals for Choosing a Model

Why public benchmarks do not transfer to your workload, how to build a model-selection eval, and why cost per successful task beats price per token.

Discussion

LLM Model LandscapeFoundational

Multimodal Input Processing

The GPT-6 tiers, what reasoning effort replaced, when to use the Responses API over Chat Completions, how the bill is calculated, and what is different on Azure.

By Amit UpadhyayPublished 13 Oct 202611 min

Step 1 / 4

Speed

What's happening

GPT-6.1 Sol Model

gpt-6.1-sol: the default GPT-6 model — text and image in, text out

💡 Analogy

Like a Swiss Army knife vs individual tools — one model handles text and images natively.

GPT-6.1 Sol is the balanced tier of the GPT-6 family: near top-tier quality at $2 input / $10 output per 1M tokens. It takes text and images and always reasons before answering.

Console Output

> model: "gpt-6.1-sol"
> Context: 1.05M tokens
> max_completion_tokens: 2000 (output + reasoning cap)
> Response mode: message content

●gpt-6.1-sol.py

GPT-6.1 Sol ModelL6–9

1from openai import OpenAI

2import base64

4client = OpenAI()

6# Text: multimodal by default, text only here

7response = client.chat.completions.create(

8 model="gpt-6.1-sol",

9 messages=[

10 {"role": "system", "content": "You are a helpful assistant."},

11 {"role": "user", "content": "Explain attention in transformers in 2 sentences."},

12 ],

13 max_completion_tokens=2000, # includes reasoning tokens

14 reasoning_effort="low", # low | medium | high | xhigh | max

15)

16print(response.choices[0].message.content)

17print(f"Tokens: {response.usage.total_tokens}")

19# Vision: pass image URL or base64

20with open("diagram.png", "rb") as f:

21 image_b64 = base64.b64encode(f.read()).decode()

23vision_response = client.chat.completions.create(

24 model="gpt-6.1-sol",

25 messages=[{

26 "role": "user",

27 "content": [

28 {"type": "text", "text": "What does this architecture diagram show?"},

29 {"type": "image_url", "image_url": {

30 "url": f"data:image/png;base64,{image_b64}",

31 "detail": "high", # high = 4096 tokens for detail

32 }},

33 ],

34 }],

35)

36print(vision_response.choices[0].message.content)

36 lines

Steps · 4

What's happening

GPT-6.1 Sol Model

gpt-6.1-sol: the default GPT-6 model — text and image in, text out

💡 Analogy

Like a Swiss Army knife vs individual tools — one model handles text and images natively.

GPT-6.1 Sol is the balanced tier of the GPT-6 family: near top-tier quality at $2 input / $10 output per 1M tokens. It takes text and images and always reasons before answering.

Console Output

> model: "gpt-6.1-sol"
> Context: 1.05M tokens
> max_completion_tokens: 2000 (output + reasoning cap)
> Response mode: message content

Next: Test →

GPT-6 OpenAI reasoning-effort Responses-API model-selection

← PreviousClaude: Anthropic's Model Family in Practice

Continue reading

All LLM Model Landscape articles →

LLM Model Landscape11 min