Enter the API URL, your customer key and a model ID in compatible software to access verified models.
https://guko.si/api/v1
Explore leading model families
Compare GPT, Claude and image models by use case and reference cost.
20 modelsUpstream reference prices · USD
gpt-5.6-sol
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-6.1-sol
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-6-sol
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-6-astra
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-5.6-terra
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-5.5
A GPT catalog version for conversation, coding assistance and writing. Evaluate each version and suffix separately.
openai-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-opus-5
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-opus-5-5
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-opus-4-8
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-opus-4-7
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-opus-4-6
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-sonnet-5-5
A Sonnet-series option for coding assistance, document workflows and conversation. Select the full model ID.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-sonnet-5
A Sonnet-series option for coding assistance, document workflows and conversation. Select the full model ID.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-sonnet-4-6
A Sonnet-series option for coding assistance, document workflows and conversation. Select the full model ID.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-haiku-4-5
A Haiku-series option for everyday questions, classification and concise replies. Evaluate it on your own workload.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-fable-5
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
claude-fable-5-1
A Claude catalog version for document analysis, problem solving and writing. Available access is shown in the console.
claude-award · 2026-10-04 snapshot. Per million tokens at the base context tier, without cache or risk adjustments.
gpt-image-2
Image generation catalog entry for illustration and visual concepts. Image access is not yet available on this platform.
openai-award · 2026-10-04 snapshot. Per-request reference only; image access is pending.
gpt-image-2.5-sunburst
Image generation catalog entry for illustration and visual concepts. Image access is not yet available on this platform.
openai-award · 2026-10-04 snapshot. Per-request reference only; image access is pending.
gpt-image-2.5-flare
Image generation catalog entry for illustration and visual concepts. Image access is not yet available on this platform.
openai-award · 2026-10-04 snapshot. Per-request reference only; image access is pending.
No matching models. Try another search.
Reference prices are from the upstream catalog, not platform retail quotes. Unified access is pending; available models, permissions and platform prices are shown in the console. Catalog names do not confirm model provenance or that every model has been tested.
Manage models, API keys, compute capacity, team access, and spend through one clear workflow.
Call multiple models through a consistent interface with centralized logs, quotas, and routing policies.
Provision capacity for inference, training, deployment, and elastic scaling as requirements change.
Control key permissions, member roles, project limits, and usage auditing from one place.
Dedicated access, private or hybrid deployment, consolidated billing, and expert technical support.
Eliminate duplicate platform work and keep your team focused on product quality and growth.
Manage access to multiple models with a single API key and integration pattern.
Choose routes based on model health, latency, capability, and cost.
Scale request throughput with clearly defined capacity and integration support.
Track requests, token consumption, spend, and call logs in one view.
Configure GPUs for training, inference, deployment, and model evaluation.
Manage team balances, quotas, and consolidated enterprise billing.
Access dedicated integration, private cloud, hybrid cloud, and deployment assistance.
Work directly with a technical team that understands local and cross-border delivery requirements.
Connect to multi-region resources and elastic scheduling options designed to improve reliability for AI applications across markets.
Locations show regions where resources may be provisioned. Availability depends on workload requirements and current capacity.
Use a familiar, OpenAI-compatible request format to reduce migration effort.
This sample illustrates the integration pattern. Production endpoints, available models, and credentials are confirmed during service onboarding.
curl https://guko.si/api/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gpt-5.6-sol",
"messages": [{"role": "user", "content": "Hello TokenForge"}]
}'Review key capabilities across APIs, GPU scheduling, accounts, billing, and technical support. Production monitoring will be connected as services go live.
Share your expected model volume, GPU requirements, concurrency, and billing preferences. We will help define a practical path to production.