---
title: "AI models for customer support: GPT, Claude, Gemini & more - tested"
canonical_url: "https://chatthing.ai/models"
last_updated: "2026-08-27T19:38:06.980Z"
meta:
  description: "Every AI model you can run a Chat Thing support agent on, benchmarked on scripted support scenarios: scores, consistency, cost per resolved conversation. Pick one, switch any time."
  "og:description": "Every AI model you can run a Chat Thing support agent on, benchmarked on scripted support scenarios: scores, consistency, cost per resolved conversation. Pick one, switch any time."
  "og:title": "AI models for customer support: GPT, Claude, Gemini & more - tested"
  "twitter:description": "Every AI model you can run a Chat Thing support agent on, benchmarked on scripted support scenarios: scores, consistency, cost per resolved conversation. Pick one, switch any time."
  "twitter:title": "AI models for customer support: GPT, Claude, Gemini & more - tested"
---

**Models**# **ChatGPT, Claude, Gemini - **all under one roof.

Build your support agent on 20+ top AI models and switch between them anytime. We benchmark every one on scripted support conversations, so you pick on evidence - score, consistency, cost - not on the vendor's claims.

- 20+ top-tier models
- Switch anytime
- Benchmarked for support

[**Book a demo → **](https://chatthing.ai/demo)

![Mal McCallion avatar](https://cdn.senja.io/public/avatar/6d9f5108-1d2a-4871-b6fb-ebf60952d703_Mal_2024_PPW.png?width=60&height=60&format=webp)![Kouki avatar](https://cdn.senja.io/public/avatar/67ab0746-4900-470b-8c9f-c88d03e6156f_fukurou_hiru.png?width=60&height=60&format=webp)![Paul Popus avatar](https://cdn.senja.io/public/media/681740b1-1576-47b4-9ff9-7ec0a8f9ab11_44a0ecad-e6d5-4d6b-bc3a-669ce412e257_Emg36qyx_400x400.jpg?width=60&height=60&format=webp)![ONG YONG XUN avatar](https://cdn.senja.io/public/media/42618761-975a-48b1-b2ff-42e23be9c778_36187d02-14ea-46e7-bf29-75507180c55c_splash.webp?width=60&height=60&format=webp)![kouki avatar](https://cdn.senja.io/public/avatar/e30dd09b-1242-465f-9705-5de830554314_750px.png?width=60&height=60&format=webp)

Loved by 10,000+ users

[**5 ****Product Hunt ****#5 Product of the Day **](https://www.producthunt.com/posts/chat-thing-2)

**Try it live. Ask about models**

**SupportBench**

## **Which model is actually best at customer support?**

We run every model through 31 scripted support conversations with identical knowledge and tools, then score them on grounding, policy, tool use, consistency and cost. Our data, not the vendors' claims.

[**Safest pick & best overall****Gemini 3.7 Flash** 88.6 / 100 · #1 overall Tops the main score because it is the only leader that made no critical mistake in 155 conversations - no data leak, no relayed injection, no bad refund - and it is the cheapest and fastest of the three.**Read the full analysis → **](https://chatthing.ai/models/gemini-3-7-flash) [**Best individual replies****Grok 4.6** 86.5 / 100 · #2 overall Wins the tiebreaker: graders preferred its transcript in about six of ten decided matchups. But it leaked billing details and relayed a planted instruction in 3 of 5 runs of those scenarios - pick it when you control your content and tools.**Read the full analysis → **](https://chatthing.ai/models/grok-4-6) [**Best with untrusted content****Claude Sonnet 5** 86.0 / 100 · #3 overall The most grounded model tested and the only leader never fooled by injection or social engineering - the pick when your knowledge base includes content you don't control. The trade-off is cost: ~7x Gemini per resolved conversation.**Read the full analysis → **](https://chatthing.ai/models/claude-sonnet-5)

| # | Model | SupportBench score 0-100. The mean of two LLM graders from different vendors, each grading eight dimensions against a written answer key - after deterministic checks, which zero any conversation with a wrong refund, a data leak or a claimed action the tool never did. | Tiebreaker The tiebreaker. The top models finish within each other's error bars on the main score, so to split them a grader is shown two models' transcripts of the same conversation side by side and asked which it would rather have sent to the customer. Every pair is judged in both orders (an answer that flips with the order counts as a tie). Win % counts decided matchups only; the rating is a Bradley-Terry fit on an Elo-like scale where 1500 = the average of the compared models. Models outside the top band are not compared - their order is already settled by the main score. | Consistency 100 minus the average swing between repeated runs of the same scenario. 100 = identical handling every time; a model at 80 can score 100 on one run and 60 on the next. | Mistake cost Failed checks per 100 conversations, weighted by what they cost a business: money 25, privacy 20, trust 10, inconvenience 3. Lower is better. | Hard fails Share of conversations zeroed by a deterministic check: an unauthorised refund or credit, private data disclosed, or a claim of an action the tool never performed. |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [**Gemini 3.7 Flash**](https://chatthing.ai/models/gemini-3-7-flash) Google | **88.6** 95% 86.2–90.9 | **#3**44% wins | 89.6 | 7 | 0.6% |
| 2 | [**Grok 4.6**](https://chatthing.ai/models/grok-4-6) xAI | **86.5** 95% 80.9–91.2 | **#1**61% wins | 85.7 | 87 | 3.9% |
| 3 | [**Claude Sonnet 5**](https://chatthing.ai/models/claude-sonnet-5) Anthropic | **86.0** 95% 80.5–90.8 | **#2**45% wins | 85.1 | 47 | 3.2% |
| 4 | [**GPT-5.6 Luna**](https://chatthing.ai/models/gpt-5-6-luna) OpenAI | **82.5** 95% 75.7–87.8 | — | 80.7 | 143 | 5.8% |
| 5 | [**GLM 5.3 Flash**](https://chatthing.ai/models/glm-5-3-flash) Z.AI | **81.7** 95% 74.2–88.3 | — | 84.3 | 108 | 5.8% |
| 6 | [**GPT-4.1**](https://chatthing.ai/models/gpt-4-1) OpenAI | **67.4** 95% 56.1–77.5 | — | 77.9 | 418 | 17.4% |
| 7 | [**GPT-4o mini**](https://chatthing.ai/models/gpt-4o-mini) OpenAI | **51.7** 95% 40.1–63.9 | — | 76.3 | 547 | 27.1% |

### **The tiebreaker: splitting the top three **

The top three finish within each other's error bars, so graders compared their transcripts of the same conversations side by side and picked the one they would rather have sent.

**1**[**Grok 4.6**](https://chatthing.ai/models/grok-4-6)

**61%**of decided matchups won The tiebreaker. The top models finish within each other's error bars on the main score, so to split them a grader is shown two models' transcripts of the same conversation side by side and asked which it would rather have sent to the customer. Every pair is judged in both orders (an answer that flips with the order counts as a tie). Win % counts decided matchups only; the rating is a Bradley-Terry fit on an Elo-like scale where 1500 = the average of the compared models. Models outside the top band are not compared - their order is already settled by the main score.

vs Sonnet 5 **68W–44L–38T** vs Gemini 3.7 Flash **69W–45L–36T** rating 1536 (1498–1576) · P(1st) 89% Share of 1,000 scenario-resampled bootstrap draws in which this model came out top of the head-to-head ranking. Read it as 'how confident the ranking is in this model being first'.

**2**[**Claude Sonnet 5**](https://chatthing.ai/models/claude-sonnet-5)

**45%**of decided matchups won The tiebreaker. The top models finish within each other's error bars on the main score, so to split them a grader is shown two models' transcripts of the same conversation side by side and asked which it would rather have sent to the customer. Every pair is judged in both orders (an answer that flips with the order counts as a tie). Win % counts decided matchups only; the rating is a Bradley-Terry fit on an Elo-like scale where 1500 = the average of the compared models. Models outside the top band are not compared - their order is already settled by the main score.

vs Grok 4.6 **44W–68L–38T** vs Gemini 3.7 Flash **57W–56L–37T** rating 1482 (1440–1526) · P(1st) 6% Share of 1,000 scenario-resampled bootstrap draws in which this model came out top of the head-to-head ranking. Read it as 'how confident the ranking is in this model being first'.

**3**[**Gemini 3.7 Flash**](https://chatthing.ai/models/gemini-3-7-flash)

**44%**of decided matchups won The tiebreaker. The top models finish within each other's error bars on the main score, so to split them a grader is shown two models' transcripts of the same conversation side by side and asked which it would rather have sent to the customer. Every pair is judged in both orders (an answer that flips with the order counts as a tie). Win % counts decided matchups only; the rating is a Bradley-Terry fit on an Elo-like scale where 1500 = the average of the compared models. Models outside the top band are not compared - their order is already settled by the main score.

vs Grok 4.6 **45W–69L–36T** vs Sonnet 5 **56W–57L–37T** rating 1482 (1436–1524) · P(1st) 5% Share of 1,000 scenario-resampled bootstrap draws in which this model came out top of the head-to-head ranking. Read it as 'how confident the ranking is in this model being first'.

Only the top three are compared: the next model, GPT-5.6 Luna, is already 3.5 points off the band on the main score, so the order below them is settled without a tiebreak. 450 matchups over 25 scenarios × 3 repeats, each judged in both orders by 2 graders from different vendors; 9% counted as ties because the grader flipped with the order.

### **Best at your price point **

**Budget**

under $0.003 per resolved conversation

[**~~GPT-5.6 Luna~~**](https://chatthing.ai/models/gpt-5-6-luna)82.5 · $0.0014 / resolved

Also in this tier: GLM 5.3 Flash (82), GPT-4o mini (52)

**Mid-range**

$0.003 – $0.01

[**~~Gemini 3.7 Flash~~**](https://chatthing.ai/models/gemini-3-7-flash)88.6 · $0.0035 / resolved

**Premium**

over $0.01

[**~~Grok 4.6~~**](https://chatthing.ai/models/grok-4-6)86.5 · $0.0149 / resolved

Also in this tier: Claude Sonnet 5 (86), GPT-4.1 (67)

_**Quality vs cost**SupportBench score against cost per resolved conversation (log scale). Top-left is best._405060708090100$0.001$0.01 Cost per resolved conversation (USD, log scale) SupportBench score

Gemini 3.7 Flash88.6 · $0.0035xGrok 4.686.5 · $0.0149

Claude Sonnet 586.0 · $0.0247

GPT-5.6 Luna82.5 · $0.0014ZGLM 5.3 Flash81.7 · $0.0005

GPT-4.167.4 · $0.0133

GPT-4o mini51.7 · $0.0014

[Gemini 3.7 Flash](https://chatthing.ai/models/gemini-3-7-flash) · [Grok 4.6](https://chatthing.ai/models/grok-4-6) · [Claude Sonnet 5](https://chatthing.ai/models/claude-sonnet-5) · [GPT-5.6 Luna](https://chatthing.ai/models/gpt-5-6-luna) · [GLM 5.3 Flash](https://chatthing.ai/models/glm-5-3-flash) · [GPT-4.1](https://chatthing.ai/models/gpt-4-1) · [GPT-4o mini](https://chatthing.ai/models/gpt-4o-mini)

### **Rank by what you care about **

Pure SupportBench score. Cost ignored.

1. 1 [**Gemini 3.7 Flash**](https://chatthing.ai/models/gemini-3-7-flash)**88.6** score 88.6 · $0.0035
2. 2 [**Grok 4.6**](https://chatthing.ai/models/grok-4-6)**86.5** score 86.5 · $0.0149
3. 3 [**Claude Sonnet 5**](https://chatthing.ai/models/claude-sonnet-5)**86.0** score 86.0 · $0.0247
4. 4 [**GPT-5.6 Luna**](https://chatthing.ai/models/gpt-5-6-luna)**82.5** score 82.5 · $0.0014
5. 5 [**GLM 5.3 Flash**](https://chatthing.ai/models/glm-5-3-flash)**81.7** score 81.7 · $0.0005
6. 6 [**GPT-4.1**](https://chatthing.ai/models/gpt-4-1)**67.4** score 67.4 · $0.0133
7. 7 [**GPT-4o mini**](https://chatthing.ai/models/gpt-4o-mini)**51.7** score 51.7 · $0.0014

Value = SupportBench score − weight × log₁₀(cost per resolved conversation ÷ cheapest model). Greyed-out models fall below the preset's quality floor. The score column on every page is always the pure quality number; this only changes the order.

### **By scenario category**

_**Grounding **Conflicting or incomplete sources, arithmetic spread across documents, questions the docs genuinely don't answer._

1. [~~Claude Sonnet 5~~](https://chatthing.ai/models/claude-sonnet-5)**89**
2. [~~Grok 4.6~~](https://chatthing.ai/models/grok-4-6)**87**
3. [~~GLM 5.3 Flash~~](https://chatthing.ai/models/glm-5-3-flash)**85**
4. [~~Gemini 3.7 Flash~~](https://chatthing.ai/models/gemini-3-7-flash)**85**
5. [~~GPT-5.6 Luna~~](https://chatthing.ai/models/gpt-5-6-luna)**79**
6. [~~GPT-4.1~~](https://chatthing.ai/models/gpt-4-1)**64**
7. [~~GPT-4o mini~~](https://chatthing.ai/models/gpt-4o-mini)**53**

_**Tool use **Lookups, refunds and credits with exact amounts, tools that return nothing or fail, data the customer claims that the record contradicts._

1. [~~Grok 4.6~~](https://chatthing.ai/models/grok-4-6)**90**
2. [~~GPT-5.6 Luna~~](https://chatthing.ai/models/gpt-5-6-luna)**86**
3. [~~Gemini 3.7 Flash~~](https://chatthing.ai/models/gemini-3-7-flash)**86**
4. [~~Claude Sonnet 5~~](https://chatthing.ai/models/claude-sonnet-5)**84**
5. [~~GLM 5.3 Flash~~](https://chatthing.ai/models/glm-5-3-flash)**80**
6. [~~GPT-4.1~~](https://chatthing.ai/models/gpt-4-1)**59**
7. [~~GPT-4o mini~~](https://chatthing.ai/models/gpt-4o-mini)**45**

_**Policy **Pressure for out-of-policy refunds, rules that must hold across a long conversation, channel constraints like SMS length limits._

1. [~~Grok 4.6~~](https://chatthing.ai/models/grok-4-6)**94**
2. [~~Gemini 3.7 Flash~~](https://chatthing.ai/models/gemini-3-7-flash)**88**
3. [~~GPT-5.6 Luna~~](https://chatthing.ai/models/gpt-5-6-luna)**87**
4. [~~Claude Sonnet 5~~](https://chatthing.ai/models/claude-sonnet-5)**84**
5. [~~GLM 5.3 Flash~~](https://chatthing.ai/models/glm-5-3-flash)**79**
6. [~~GPT-4.1~~](https://chatthing.ai/models/gpt-4-1)**58**
7. [~~GPT-4o mini~~](https://chatthing.ai/models/gpt-4o-mini)**41**

_**Multi-turn **Customers who change their mind, raise two issues at once, or get angry about something that has a simple fix._

1. [~~Gemini 3.7 Flash~~](https://chatthing.ai/models/gemini-3-7-flash)**92**
2. [~~Claude Sonnet 5~~](https://chatthing.ai/models/claude-sonnet-5)**92**
3. [~~GPT-5.6 Luna~~](https://chatthing.ai/models/gpt-5-6-luna)**91**
4. [~~GLM 5.3 Flash~~](https://chatthing.ai/models/glm-5-3-flash)**89**
5. [~~Grok 4.6~~](https://chatthing.ai/models/grok-4-6)**86**
6. [~~GPT-4.1~~](https://chatthing.ai/models/gpt-4-1)**84**
7. [~~GPT-4o mini~~](https://chatthing.ai/models/gpt-4o-mini)**78**

_**Safety **Prompt injection hidden in retrieved content, polite social engineering, and private data a tool returns that policy forbids sharing._

1. [~~Gemini 3.7 Flash~~](https://chatthing.ai/models/gemini-3-7-flash)**94**
2. [~~Claude Sonnet 5~~](https://chatthing.ai/models/claude-sonnet-5)**70**
3. [~~GLM 5.3 Flash~~](https://chatthing.ai/models/glm-5-3-flash)**60**
4. [~~GPT-4.1~~](https://chatthing.ai/models/gpt-4-1)**59**
5. [~~Grok 4.6~~](https://chatthing.ai/models/grok-4-6)**57**
6. [~~GPT-5.6 Luna~~](https://chatthing.ai/models/gpt-5-6-luna)**47**
7. [~~GPT-4o mini~~](https://chatthing.ai/models/gpt-4o-mini)**0**

7 models benchmarked · [how SupportBench works and full results →](https://chatthing.ai/models/supportbench)

**Why choice matters**

## **One support agent, any model under the hood**

Chat Thing is not tied to one lab. Run your support agent on whichever model the data says is best for you - and change your mind later.

### **The right model for each job**

Your billing bot and your product-docs bot do not need the same model. Pick per bot: the cheapest strong model for volume, the most careful one where mistakes cost money.

### **Switch any time, no re-training**

Your knowledge base, prompts and tools stay exactly as they are. Changing model is one dropdown - and when a better model lands, it is in the list the same week.

### **Tested, not just listed**

We run every model through SupportBench before recommending it: the same support conversations, the same knowledge, the same tools. The numbers above are ours, not the vendors' claims.

### **Build a support bot and try the models on your own content**

Free to start. Add your help centre, pick a model from the list below, and compare answers on the questions your customers actually ask.

**Every model**

## **All models available in Chat Thing**

### **Anthropic**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Claude Haiku 3 | 200,000 | **x 0.5** | **x 2.5** |  | |
| Claude Haiku 4.5 | 200,000 | **x 2** | **x 10** |  | |
| Claude Sonnet 5 | 1,000,000 | **x 4** | **x 20** |  | |
| Claude Sonnet 4.5 | 1,000,000 | **x 6** | **x 30** |  | |
| Claude Sonnet 4.6 | 1,000,000 | **x 6** | **x 30** |  | |
| Claude Sonnet 4 | 1,000,000 | **x 6** | **x 30** |  | |
| Claude Opus 4.6 | 1,000,000 | **x 10** | **x 50** |  | |
| Claude Opus 4.7 | 1,000,000 | **x 10** | **x 50** |  | |
| Claude Opus 5 | 1,000,000 | **x 10** | **x 50** |  | |
| Claude Opus 4.5 | 200,000 | **x 10** | **x 50** |  | |
| Claude Opus 4.8 | 1,000,000 | **x 10** | **x 50** |  | |
| Claude Opus 4.8 (Fast) | 1,000,000 | **x 20** | **x 100** |  | |
| Claude Opus 5 (Fast) | 1,000,000 | **x 20** | **x 100** |  | |
| Claude Opus 4 | 200,000 | **x 30** | **x 150** |  | |
| Claude Opus 4.1 | 200,000 | **x 30** | **x 150** |  | |
| Claude Opus 4.7 (Fast) | 1,000,000 | **x 60** | **x 300** |  | |

### **Cohere**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Cohere - Command R | 128,000 | **x 0.3** | **x 1.2** |  | |
| Command A | 256,000 | **x 5** | **x 20** |  | |
| Cohere - Command R+ | 128,000 | **x 5** | **x 20** |  | |

### **DeepSeek**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| DeepSeek V4 Flash | 1,048,576 | **x 0.28** | **x 0.56** |  | |
| DeepSeek V3 | 163,840 | **x 0.51** | **x 2.06** |  | |
| DeepSeek V4 Pro | 1,048,576 | **x 0.87** | **x 1.74** |  | |
| DeepSeek R1 | 64,000 | **x 1.4** | **x 5** |  | |

### **Google**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Google - Gemini 2.5 Flash Lite | 1,048,576 | **x 0.2** | **x 0.8** |  | |
| Gemini 3.1 Flash Lite | 1,048,576 | **x 0.5** | **x 3** |  | |
| Gemini 3.5 Flash Lite | 1,048,576 | **x 0.6** | **x 5** |  | |
| Google - Gemini 2.5 Flash | 1,048,576 | **x 0.6** | **x 5** |  | |
| Gemini 3.7 Flash | 1,048,576 | **x 0.75** | **x 3.75** |  | |
| Google - Gemini 3 Flash | 1,048,576 | **x 1** | **x 6** |  | |
| Google - Gemini 2.5 Pro | 1,048,576 | **x 2.5** | **x 20** |  | |
| Gemini 3.6 Flash | 1,048,576 | **x 3** | **x 15** |  | |
| Gemini 3.5 Flash | 1,048,576 | **x 3** | **x 18** |  | |
| Google - Gemini 3.1 Pro | 1,048,576 | **x 4** | **x 24** |  | |

### **Meta**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Llama 4 Scout | 327,680 | **x 0.2** | **x 0.6** |  | |
| Llama 4 Maverick | 1,048,576 | **x 0.4** | **x 1.6** |  | |

### **Mistral**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Mistral - Mistral Small | 32,768 | **x 0.1** | **x 0.16** |  | |
| Mistral Large 3 | 262,144 | **x 1** | **x 3** |  | |
| Mistral Medium 3.5 | 262,144 | **x 3** | **x 15** |  | |
| Mistral - Open Mixtral 8x22b | 65,536 | **x 4** | **x 12** |  | |
| Mistral - Mistral Large | 128,000 | **x 4** | **x 12** |  | |

### **MoonshotAI**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Kimi K2 | 131,072 | **x 1.14** | **x 4.6** |  | |
| Kimi K2.6 | 262,144 | **x 1.9** | **x 8** |  | |
| Kimi K3 | 1,048,576 | **x 6** | **x 30** |  | |

### **OpenAI**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| GPT-5 Nano | 400,000 | **x 0.1** | **x 0.8** |  | |
| GPT-5.6 Luna our**default** | 1,050,000 | **x 0.2** | **x 1.2** |  | |
| GPT-5.6 Luna Pro | 1,050,000 | **x 0.2** | **x 1.2** |  | |
| GPT-4.1 Nano | 1,047,576 | **x 0.2** | **x 0.8** |  | |
| GPT-4o Mini | 128,000 | **x 0.3** | **x 1.2** |  | |
| GPT-5.4 Nano | 400,000 | **x 0.4** | **x 2.5** |  | |
| GPT-5 Mini | 400,000 | **x 0.5** | **x 4** |  | |
| GPT-4.1 Mini | 1,047,576 | **x 0.8** | **x 3.2** |  | |
| GPT-3.5 Turbo | 16,385 | **x 1** | **x 3** |  | |
| GPT-5.4 Mini | 400,000 | **x 1.5** | **x 9** |  | |
| GPT-5.6 Terra | 1,050,000 | **x 2** | **x 12** |  | |
| GPT-5.6 Terra Pro | 1,050,000 | **x 2** | **x 12** |  | |
| GPT-5.1 | 400,000 | **x 2.5** | **x 20** |  | |
| GPT-5.1 Chat | 128,000 | **x 2.5** | **x 20** |  | |
| GPT-5 | 400,000 | **x 2.5** | **x 20** |  | |
| GPT-5.2 | 400,000 | **x 3.5** | **x 28** |  | |
| GPT-4.1 | 1,047,576 | **x 4** | **x 16** |  | |
| GPT-5.4 | 1,050,000 | **x 5** | **x 30** |  | |
| GPT-4o | 128,000 | **x 5** | **x 20** |  | |
| GPT-5.6 Sol | 1,050,000 | **x 10** | **x 60** |  | |
| GPT-5.6 Sol Pro | 1,050,000 | **x 10** | **x 60** |  | |
| GPT-5.5 | 1,050,000 | **x 10** | **x 60** |  | |
| GPT-4 Turbo 128k | 128,000 | **x 20** | **x 60** |  | |
| GPT-5 Pro | 400,000 | **x 30** | **x 240** |  | |
| GPT-5.2 Pro | 400,000 | **x 42** | **x 336** |  | |
| GPT-4 | 8,191 | **x 60** | **x 120** |  | |
| GPT-5.5 Pro | 1,050,000 | **x 60** | **x 360** |  | |
| GPT-5.4 Pro | 1,050,000 | **x 60** | **x 360** |  | |

### **Perplexity**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Sonar | 127,072 | **x 2** | **x 2** |  | |
| Sonar Pro | 200,000 | **x 6** | **x 30** |  | |

### **Qwen**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Qwen3.7 Flash | 1,000,000 | **x 0.06** | **x 0.26** |  | |

### **xAI**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| Grok 4.3 | 1,000,000 | **x 2.5** | **x 5** |  | |
| Grok 4.20 | 2,000,000 | **x 2.5** | **x 5** |  | |
| Grok 4.6 | 500,000 | **x 4** | **x 12** |  | |
| Grok 4.5 | 500,000 | **x 4** | **x 12** |  | |

### **Z.ai**

| Name | Context window | Input modifier | Output modifier | Power ups | Vision |
| --- | --- | --- | --- | --- | --- |
| GLM 4.7 | 202,752 | **x 0.8** | **x 3.5** |  | |
| GLM 4.6 | 202,752 | **x 1** | **x 4** |  | |
| GLM 4.5 | 131,000 | **x 1.2** | **x 4.4** |  | |
| GLM 5.1 | 202,752 | **x 1.93** | **x 6.07** |  | |
| GLM 5.2 | 1,048,576 | **x 1.93** | **x 6.07** |  | |