Model Command Center

Operational benchmarks, pricing, and capability matrix for 15 tracked models.

Cost & Workload Calculator

— estimate monthly API spend

Model

Input Tokens per Request (M)

Output Tokens per Request (M)

Requests / Day

Select a model to see cost estimates

Operational performance

— verified latency and throughput

Compare time-to-first-token, p50 latency, throughput, and builder decision tags when verified provider or test-lab measurements are available.

Model	TTFT	P50 latency	Throughput	Decision tags
GPT-5.6 Sol API model	—	—	—	Awaiting verified measurement
GPT-5.6 Terra API model	—	—	—	Awaiting verified measurement
GPT-5.6 Luna API model	—	—	—	Awaiting verified measurement
Claude Fable 5 API model	—	—	—	Awaiting verified measurement
Gemini 3.7 Flash API model	—	—	—	Awaiting verified measurement
Gemini 1.5 Flash API model	—	—	—	Awaiting verified measurement
Llama 3.1 405B Open weights	—	—	—	Awaiting verified measurement
GPT-4o API model	—	—	—	Awaiting verified measurement
Mistral Large API model	—	—	—	Awaiting verified measurement
DeepSeek V3 Open weights	—	—	—	Awaiting verified measurement
Llama 3.1 70B Open weights	—	—	—	Awaiting verified measurement
GPT-4o Mini API model	—	—	—	Awaiting verified measurement
Claude 3.5 Sonnet API model	—	—	—	Awaiting verified measurement
Claude 3 Haiku API model	—	—	—	Awaiting verified measurement
Gemini 1.5 Pro API model	—	—	—	Awaiting verified measurement

Measurements are kept separate from model-provider claims and labeled only after source verification.

Filters

Open Weights Only

Min Context

Max Input Price ($/M)

15 / 15 models

Model	Family	Weights	Context	Input/M	Output/M	Benchmarks	Builder Notes
Claude 3 Haiku Proprietary API	Claude 3	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Claude 3.5 Sonnet Proprietary API	Claude 3	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Claude Fable 5 Proprietary API	Claude 5	Closed	—	$10.00	$50.00	—	Standard API list pricing. Cache reads, cache writes, Batch API, Fast mode, and US-only inference can change the effective cost.
DeepSeek V3 Open weights; verify current license	DeepSeek V3	Open	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
GPT-4o Proprietary API	GPT-4o	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
GPT-4o Mini Proprietary API	GPT-4o	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
GPT-5.6 Luna Proprietary API	GPT-5.6	Closed	—	$0.20	$1.20	—	Standard short-context list pricing. Cached-input, long-context, Batch API, and Fast-mode pricing are different; verify workload-specific rates on the provider page.
GPT-5.6 Sol Proprietary API	GPT-5.6	Closed	—	$5.00	$30.00	—	Standard short-context list pricing. Cached-input, long-context, Batch API, and Fast-mode pricing are different; verify workload-specific rates on the provider page.
GPT-5.6 Terra Proprietary API	GPT-5.6	Closed	—	$2.00	$12.00	—	Standard short-context list pricing. Cached-input, long-context, Batch API, and Fast-mode pricing are different; verify workload-specific rates on the provider page.
Gemini 1.5 Flash Proprietary API	Gemini 1.5	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Gemini 1.5 Pro Proprietary API	Gemini 1.5	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Gemini 3.7 Flash Proprietary API	Gemini 3.7	Closed	—	$0.75	$3.75	—	Paid-tier introductory list pricing through 2026-12-31; Google lists $1.50 input and $7.50 output beginning 2027-01-01. Free-tier pricing is not represented in this tracker row.
Llama 3.1 405B Llama Community License	Llama 3.1	Open	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Llama 3.1 70B Llama Community License	Llama 3.1	Open	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.
Mistral Large Commercial API	Mistral Large	Closed	—	—	—	—	Awaiting current source-verified pricing, context, and benchmark metadata before public SEO indexing.