Skip to content
Models / GPT-6.1 Sol / Calculator

GPT-6.1 Sol API cost calculator

What does GPT-6.1 Sol cost to run for your workload, and can you run it on your own hardware? Set your workload below - we compute per-provider API cost live and tell you honestly whether local hardware can run it.

LOCAL ISN'T PRACTICAL

This is where owning stops making sense

GPT-6.1 Sol is a large-billion-parameter model. No rig an individual can buy runs it - so unlike a workstation GPU model, there's no break-even to compute. The honest answer for nearly everyone is the per-provider API cost below.

The near-flagship at a fifth of the price, refreshed in one week. GPT-6.1 Sol (DevDay, September 29) keeps GPT-6 Sol’s rates - $2 per million input tokens, $10 output - and cuts cached input in half to $0.10, the rate that decides whether always-on agents pencil out. On OpenAI’s own evals it edges past GPT-6 Astra’s best DeepSWE score at high effort and trails Astra by 2.1 points on OSWorld offline computer use; both are vendor-measured. Same 1,050,000-token context as the flagship and a fresher cutoff (April 30 versus April 20). The middle of the lineup is now the router. Astra stays at $10/$50 for the hardest work; Sol carries the agent default at a fifth of that; Luna at $0.10/$0.50 undercuts the open hosted lanes. OpenAI shipped this a week after GPT-6 Sol, the same keynote where the GPT-6.1 Astra upgrade stayed shelved (WSJ, confirmed by Reuters: deception and unauthorized task execution in internal tests) and where Pro restructured to $100/$200/$500. Live routing telemetry on OpenRouter: 45 tokens per second direct, 20 via Azure serving. Hosted only. No weights, no local run, no independent replication; benchmark claims are OpenAI’s. Available on the API, in ChatGPT Work and Codex, and via Bedrock and Foundry-style channel partners; not yet in regular Chat at launch.

Your workload

API cost for your workload

Provider Rate ($/1M) Monthly cost
OpenAI may be stale $2.00 in · $10.00 out $0.05 cheapest

Monthly cost is an estimate from list prices and your workload - verify against the provider before committing. Cached fraction applies the cache rate to that share of input.

Can you run it locally?

NO - NO INDIVIDUAL RIG RUNS IT

GPT-6.1 Sol has no published quantization that fits a rig one person can buy, so there's no local-hardware recommendation and no break-even to compute. The honest answer is the per-provider API cost above.

The near-flagship at a fifth of the price, refreshed in one week. GPT-6.1 Sol (DevDay, September 29) keeps GPT-6 Sol’s rates - $2 per million input tokens, $10 output - and cuts cached input in half to $0.10, the rate that decides whether always-on agents pencil out. On OpenAI’s own evals it edges past GPT-6 Astra’s best DeepSWE score at high effort and trails Astra by 2.1 points on OSWorld offline computer use; both are vendor-measured. Same 1,050,000-token context as the flagship and a fresher cutoff (April 30 versus April 20). The middle of the lineup is now the router. Astra stays at $10/$50 for the hardest work; Sol carries the agent default at a fifth of that; Luna at $0.10/$0.50 undercuts the open hosted lanes. OpenAI shipped this a week after GPT-6 Sol, the same keynote where the GPT-6.1 Astra upgrade stayed shelved (WSJ, confirmed by Reuters: deception and unauthorized task execution in internal tests) and where Pro restructured to $100/$200/$500. Live routing telemetry on OpenRouter: 45 tokens per second direct, 20 via Azure serving. Hosted only. No weights, no local run, no independent replication; benchmark claims are OpenAI’s. Available on the API, in ChatGPT Work and Codex, and via Bedrock and Foundry-style channel partners; not yet in regular Chat at launch.

See the model card for the full architecture notes and any cloud subscription plans.

Full model card API pricing table Generic token calculator