Goatfied routes each task to the right model, with automatic fallbacks. Here's the full table.
The models
Every model below is served directly and included with your plan — our own GOAT models alongside frontier third-party models (Claude, GPT, Gemini, Grok, Llama and more). Rates are token-metered, shown per 1M tokens. The table is generated from the same live catalog the in-app model picker reads, so newly released models appear here automatically.
87 models across 13 providersAuto our GOAT / Composer modelsAPI third-party, billed at API rates
GOAT
4 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Auto· tools· vision | Auto | $1.25 | $0.25 | $6 | 200K |
| GOAT - Free· tools· vision | Auto | $0.50 | $0.20 | $2.50 | 200K |
| GOAT - Fast· tools· vision | Auto | $0.50 | $0.20 | $2.50 | 200K |
| GOAT - Frontier· tools· vision | Auto | $0.50 | $0.20 | $2.50 | 200K |
Anthropic
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Claude Opus 5.5· tools· vision | API | $4 | $0.40 | $20 | 1M |
| Claude Opus 5.5 (batch)· tools· vision | API | $2 | $0.20 | $10 | 1M |
| Claude Fable 5.1· tools· vision | API | $10 | $1 | $50 | 1M |
| Claude Fable 5.1 (batch)· tools· vision | API | $5 | $0.50 | $25 | 1M |
| Claude Opus 5· tools· vision | API | $5 | $0.50 | $25 | 1M |
| Claude Opus 5 (batch)· tools· vision | API | $2.50 | $0.25 | $12.50 | 1M |
| Claude Sonnet 5· tools· vision | API | $2 | $0.20 | $10 | 1M |
| Claude Sonnet 5 (batch)· tools· vision | API | $1 | $0.10 | $5 | 1M |
OpenAI
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| GPT-6 Luna Pro· tools· vision | API | $0.10 | $0.01 | $0.50 | 1.1M |
| GPT-6 Luna Pro (batch)· tools· vision | API | $0.05 | $0.01 | $0.25 | 1.1M |
| GPT-6 Luna· tools· vision | API | $0.10 | $0.01 | $0.50 | 1.1M |
| GPT-6 Luna (batch)· tools· vision | API | $0.05 | $0.01 | $0.25 | 1.1M |
| GPT-6 Sol Pro· tools· vision | API | $2 | $0.20 | $10 | 1.1M |
| GPT-6 Sol Pro (batch)· tools· vision | API | $1 | $0.10 | $5 | 1.1M |
| GPT-6 Sol· tools· vision | API | $2 | $0.20 | $10 | 1.1M |
| GPT-6 Sol (batch)· tools· vision | API | $1 | $0.10 | $5 | 1.1M |
xAI
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Grok 4.7· tools· vision | API | $1.60 | $0.16 | $4.80 | 500K |
| Grok 4.6· tools· vision | API | $2 | $0.20 | $6 | 500K |
| Grok 4.5· tools· vision | API | $2 | $0.20 | $6 | 500K |
| Grok Build 0.1· tools· vision | API | $1 | $0.10 | $2 | 256K |
| Grok 4.3· tools· vision | API | $1.25 | $0.13 | $2.50 | 1M |
| Grok 4.3 (batch)· tools· vision | API | $1 | $0.10 | $2 | 1M |
| Grok 4.20 Multi-Agent· vision | API | $1.25 | $0.13 | $2.50 | 2M |
| Grok 4.20· tools· vision | API | $1.25 | $0.13 | $2.50 | 2M |
Google
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Gemini 3.8 Flash· tools· vision | API | $0.75 | $0.07 | $3.75 | 1.0M |
| Gemini 3.8 Flash (batch)· tools· vision | API | $0.38 | $0.04 | $1.88 | 1.0M |
| Gemini 3.7 Flash· tools· vision | API | $0.75 | $0.07 | $3.75 | 1.0M |
| Gemini 3.7 Flash (batch)· tools· vision | API | $0.38 | $0.04 | $1.88 | 1.0M |
| Gemini 3.6 Flash· tools· vision | API | $0.75 | $0.07 | $3.75 | 1.0M |
| Gemini 3.6 Flash (batch)· tools· vision | API | $0.38 | $0.04 | $1.88 | 1.0M |
| Gemini 3.5 Flash Lite· tools· vision | API | $0.30 | $0.03 | $2.50 | 1.0M |
| Gemini 3.5 Flash Lite (batch)· tools· vision | API | $0.15 | $0.01 | $1.25 | 1.0M |
Meta
7 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Llama 4 Maverick· tools· vision | API | $0.19 | $0.02 | $0.65 | 1.0M |
| Llama 4 Scout· tools· vision | API | $0.10 | $0.01 | $0.30 | 1.3M |
| Llama 3.3 70B Instruct· tools | API | $0.10 | $0.01 | $0.32 | 131K |
| Llama 3.2 1B Instruct | API | $0.03 | $0.00 | $0.20 | 60K |
| Llama 3.2 3B Instruct | API | $0.05 | $0.01 | $0.33 | 131K |
| Llama 3.1 70B Instruct· tools | API | $0.40 | $0.04 | $0.40 | 131K |
| Llama 3.1 8B Instruct· tools | API | $0.05 | $0.01 | $0.08 | 131K |
DeepSeek
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| DeepSeek V4.1 Flash· tools· vision | API | $0.30 | $0.03 | $1.20 | 1.0M |
| DeepSeek V4.1 Flash (batch)· tools· vision | API | $0.11 | $0.01 | $0.34 | 1.0M |
| DeepSeek V4 Pro 0813· tools | API | $0.26 | $0.03 | $0.79 | 1.0M |
| DeepSeek V4 Flash 0731· tools | API | $0.02 | $0.00 | $0.32 | 1.3M |
| DeepSeek V4 Pro 0423· tools | API | $0.42 | $0.04 | $0.84 | 1.0M |
| DeepSeek V4 Flash 0423· tools | API | $0.05 | $0.00 | $0.09 | 1.0M |
| DeepSeek V3.2· tools | API | $0.27 | $0.03 | $0.40 | 164K |
| DeepSeek V3.2 Exp· tools | API | $0.27 | $0.03 | $0.41 | 164K |
Mistral
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Mistral Medium 3.5· tools· vision | API | $1.50 | $0.15 | $7.50 | 262K |
| Mistral Medium 3.5 (batch)· tools· vision | API | $0.75 | $0.07 | $3.75 | 262K |
| Mistral Small 4· tools· vision | API | $0.15 | $0.01 | $0.60 | 262K |
| Mistral Small 4 (batch)· tools· vision | API | $0.07 | $0.01 | $0.30 | 262K |
| Devstral 2 2512· tools | API | $0.40 | $0.04 | $2 | 262K |
| Ministral 3 14B 2512· tools· vision | API | $0.20 | $0.02 | $0.20 | 262K |
| Ministral 3 8B 2512· tools· vision | API | $0.15 | $0.01 | $0.15 | 262K |
| Ministral 3 8B 2512 (batch)· tools· vision | API | $0.07 | $0.01 | $0.07 | 262K |
Qwen
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Qwen3.8 Max Prime· tools· vision | API | $4 | $0.40 | $12 | 1M |
| Qwen3.8 Omni Flash· tools· vision | API | $0.15 | $0.01 | $0.47 | 1M |
| Qwen3.8 Max (0902)· tools· vision | API | $2 | $0.20 | $6 | 1M |
| Qwen3.8 Flash· tools· vision | API | $0.15 | $0.01 | $0.47 | 1M |
| Qwen3.8 27B· tools· vision | API | $0.42 | $0.04 | $3 | 1M |
| Qwen3.8 2.4T A95B· tools | API | $2 | $0.20 | $6 | 1.0M |
| Qwen3.7 Flash· tools· vision | API | $0.03 | $0.00 | $0.13 | 1M |
| Qwen3.7 Plus· tools· vision | API | $0.32 | $0.03 | $1.28 | 1M |
Moonshot
8 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Kimi K3· tools· vision | Auto | $0.50 | $0.20 | $2.50 | 1.0M |
| Kimi K3 (batch)· tools· vision | API | $2.28 | $0.23 | $11.40 | 1.0M |
| Kimi K2.7 Code· tools· vision | API | $0.66 | $0.07 | $3.30 | 262K |
| Kimi K2.6· tools· vision | API | $0.95 | $0.10 | $4 | 262K |
| Kimi K2.5· tools· vision | API | $0.45 | $0.04 | $2.25 | 262K |
| Kimi K2 Thinking· tools | API | $0.60 | $0.06 | $2.50 | 262K |
| Kimi K2 0905· tools | API | $0.60 | $0.06 | $2.50 | 262K |
| Kimi K2 0711· tools | API | $0.57 | $0.06 | $2.30 | 131K |
Cohere
5 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Command A+· tools· vision | API | $0.30 | $0.03 | $1.50 | 192K |
| Command A | API | $2.50 | $0.25 | $10 | 256K |
| Command R7B (12-2024) | API | $0.04 | $0.00 | $0.15 | 128K |
| Command R (08-2024)· tools | API | $0.15 | $0.01 | $0.60 | 128K |
| Command R+ (08-2024)· tools | API | $2.50 | $0.25 | $10 | 128K |
Amazon
5 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Nova 2 Lite· tools· vision | API | $0.30 | $0.03 | $2.50 | 1M |
| Nova Premier 1.0· tools· vision | API | $2.50 | $0.25 | $12.50 | 1M |
| Nova Lite 1.0· tools· vision | API | $0.06 | $0.01 | $0.24 | 300K |
| Nova Micro 1.0· tools | API | $0.04 | $0.00 | $0.14 | 128K |
| Nova Pro 1.0· tools· vision | API | $0.80 | $0.08 | $3.20 | 300K |
Microsoft
2 models| Model | Pool | Input $/M | Cache read $/M | Output $/M | Context |
|---|
| Phi 4 | API | $0.07 | $0.01 | $0.14 | 16K |
| WizardLM-2 8x22B | API | $0.62 | $0.06 | $0.62 | 66K |
Rates are USD per 1M tokens. This table is generated from the same live catalog the in-app model picker reads, so newly released models appear automatically.
Plans
Every plan includes two usage pools that reset each cycle: an Auto pool for our GOAT / Composer models and an API pool for third-party models at their token rates above. When a pool’s allowance is used up you can continue on-demand at the same rates, or upgrade.
| Plan | Price | Included Auto usage | Included API usage |
|---|
| Hobby | Free | $5 | Free |
| Pro | $20/mo | $40 | $20 |
| Pro+ | $60/mo | $140 | $70 |
| Ultra | $200/mo | $800 | $400 |
| Teams Standard | $40/mo · seat | $80 / seat | $40 / seat |
| Teams Premium | $120/mo · seat | $400 / seat | $200 / seat |
| Enterprise | Contact | Unlimited | Unlimited |
See pricing for full plan details and plans & usage for how the pools meter.
Third-party models, served directly
You don’t need your own API keys — frontier third-party models are served directly and metered by tokens from the API pool, exactly like our GOAT models. Prefer your own provider account? Bring your own key or wire a custom endpoint.
Context windows
Context windows vary by model; the picker shows the limit for the one you have selected. Goatfied pairs them with codebase retrieval so relevant files are fetched rather than dumped wholesale into the prompt.
Routing & fallbacks
Each surface picks a default model, and you can override it per task in the model picker. If a model is rate-limited or unavailable, the request falls back to an equivalent model so your work does not stall.
