OpenAI's and Google's mid-tier models each have a pricing condition to plan around: a long-context band for GPT-6.1 Sol, and an intro rate for Gemini 3.8 Flash that ends December 31, 2026.
Gemini 3.8 Flash is about 2.7× cheaper than GPT-6.1 Sol on standard prompts through December 31. From January 1 Flash's price doubles, cutting that to about 1.3×. On prompts above 272K tokens, Sol's rate rises and Flash's doesn't, so long-context work keeps a bigger gap.
| GPT-6.1 Sol | Gemini 3.8 Flash | |
|---|---|---|
| Provider | OpenAI | |
| Tier | Mid-tier | Mid-tier |
| Input / 1M tokens | $2.00 | $0.75 |
| Output / 1M tokens | $10.00 | $3.75 |
| Cache hit / 1M | $0.10 | $0.075 |
| Cache write / 1M | $2.50 | — |
| Context window | 1M | 1M |
| Notes | New (Sep 29, 2026 DevDay), below Astra in the GPT-6 family; replaces GPT-6 Sol at the same input/output price with a 50% cheaper cache-read rate; >272K tokens: $4/$15, cache $0.20. ChatGPT Work/Codex on Plus/Pro/Business/Enterprise/Edu, plus API | On the Gemini API free and paid tiers; Gemini 3.7 Flash retired Oct 8, 2026 and now auto-routes here (requests still billed at this rate); intro price thru Dec 31, reverts to $1.50/$7.50 (cache $0.15) Jan 1 |
Gemini 3.8 Flash is 2.7× cheaper on input tokens ($0.75 vs $2.00 per 1M). Gemini 3.8 Flash is 2.7× cheaper on output tokens ($3.75 vs $10.00 per 1M).
| Workload | GPT-6.1 Sol | Gemini 3.8 Flash | Difference |
|---|---|---|---|
| Support chatbot 10,000 conversations × (2,000 input + 400 output tokens) | $80.00 | $30.00 | Gemini 3.8 Flash saves $50.00 |
| Document Q&A 1,000 questions × (30,000 input + 800 output tokens) | $68.00 | $25.50 | Gemini 3.8 Flash saves $42.50 |
| Coding agent session 4M input tokens (90% cache hits) + 150,000 output tokens | $2.66 | $1.13 | Gemini 3.8 Flash saves $1.53 |
Standard rates per 1M tokens from data on this site. The coding-agent row assumes 90% of input is read from the prompt cache (full input price where a model has no cache rate). Cache-write fees, batch discounts and long-context surcharges listed in the notes are not included. Updated November 2, 2026.
you're on OpenAI's platform, your prompts stay under 272K tokens, and you can use the half-price Flex or Batch tiers or OpenAI's long-lived caches.
you want the cheaper per-token rate, send long prompts, or want a free API tier for prototyping.
Sol's requests over 272K tokens cost 2× input and 1.5× output. Flash's whole window is one price, but that price doubles on January 1, 2027. After January, a 100K-token prompt is about 1.3× cheaper on Flash; a 400K-token prompt is about 2× to 2.7× cheaper, depending on the input/output mix.
Sol charges $2.50 per 1M tokens to write a 30-minute cache (a 12-hour cache costs more) and $0.10 to read it — half GPT-6 Sol's old cache-read rate. Flash's cache hit is $0.075 through December, with an hourly storage charge for cached content. Long-lived caches on Sol avoid storage fees but pay a higher write.
OpenAI's Batch and Flex tiers halve Sol's price; Google's Batch API halves Flash's. Google also lists a free tier for Gemini 3.8 Flash. Note that Gemini 3.7 Flash was retired on October 8, 2026 and now routes to 3.8 Flash at 3.8 Flash's rate.
Yes: $0.75 vs $2.00 input and $3.75 vs $10.00 output per 1M tokens through December 31, 2026. It stays cheaper after its January 1 price increase, by a smaller margin.
Yes, requests over 272K tokens are billed at a higher input, output and cache-hit rate.
GPT-6.1 Sol, released at OpenAI DevDay on September 29, 2026, at the same input and output price with a 50% cheaper cache-read rate.
How we choose: prices come from each vendor's official pricing page and are re-checked every week. We take no affiliate fees, so rankings aren't for sale.