Every monthly plan that includes GPT-6 Sol, converted to requests per $1 at the model's typical request shape. Same engine as the
main calculator
— proxy estimates carry a ~ marker; coarse estimates carry ≈.
GPT-6 Sol (API gpt-6-sol) is OpenAI's GPT-6 mid-tier for complex coding and agentic work — released 2026-09-22 at $2/$10 per 1M tokens (cached input $0.20, cache writes $2.50) beneath the GPT-6 Astra flagship, with a 1.05M-token context window.
At GPT-6 Sol’s typical shape
~153req / $1
on LLM Gateway DevPass Pro · 18 plans priced · proxy
Claude Opus 5.5Claude Opus 5GPT-6.1 SolGPT-6 SolGPT-6 AstraGPT-5.6 SolGemini 4 ArgonGemini 3.8 FlashGemini 3.7 FlashClaude Sonnet 5.5Claude Sonnet 5Claude Sonnet 4.6Claude Sonnet 4Claude Fable 5.1Claude Fable 5Claude Haiku 5.5Claude Haiku 4.5Claude Opus 4.8Claude Opus 4.8 (fast mode) (Preview)GPT-6 LunaGPT-5.6 TerraGPT-5.6 LunaGPT-5.5GPT-5.4GPT-5.4 MiniGPT-5.3-CodexGPT-5.2-CodexGPT-5.2GPT-5 MiniMAI-Code-1.1-FlashGrok 4.7Grok 4.6Grok 4.5Kimi K3
Ceiling
$70/mo total AI credits · 4× Pro usage
Reachability
—
Pricing notes
Annual Pro+ plans bought before 2026-06-01 stay on legacy premium-request billing (premium requests, not AI credits) with a frozen model set — no new models or features until the term ends, then the subscriber drops to Copilot Free; the $70/mo AI-credit ceiling describes the monthly plan (docs.github.com en/copilot/reference/copilot-billing/request-based-billing-legacy, read 2026-10-23)
GPT-6 AstraGPT-6.1 SolGPT-6 SolGPT-6 LunaGPT-5.6 Sol ProGPT-5.6 SolGPT-5.6 TerraGPT-5.6 Luna
Ceiling
More included usage than Pro 100, less than Pro 500
Reachability
—
Pricing notes
All Pro 200 subscriptions carry the same Work and Codex allowance · Pro plans have no five-hour limit; OpenAI publishes local-message estimates for Plus and Standard Business only · local messages and cloud chats share the allowance, and weekly limits may also apply · Pro 200 additionally returns $200/mo in OpenAI API credits on a linked API organization — a separate API budget (prorated first grant, expires at the end of the billing cycle, no rollover) that excludes Codex usage through API keys and does not raise ChatGPT or Codex usage limits · Third-party apps use the plan through Sign in with ChatGPT (per-app weekly cap; no API key)
Cost rule
≈coarse estimate · platform.openai.com/docs/pricing — gpt-6-sol row ($2/$10, cached input $0.20, cache write $2.50; verified 2026-09-23); Codex availability: developers.openai.com/codex/pricing lists GPT-6 Sol at 15-150 local messages per 5h on its Plus and Standard Business table and states Pro plans currently have no five-hour limit; it publishes no Pro message estimate (read 2026-10-06)
OpenAIPro 500Highest included usage of the three Pro tiers
GPT-6 AstraGPT-6.1 SolGPT-6 SolGPT-6 LunaGPT-5.6 Sol ProGPT-5.6 SolGPT-5.6 TerraGPT-5.6 Luna
Ceiling
Highest included usage of the three Pro tiers
Reachability
—
Pricing notes
Astra Ultrafast is included only on Pro 500 among the Pro tiers ($100 / $200 / $500) — it draws the included usage allowance first, then credits, and buying credits on Pro 100 or Pro 200 does not unlock it · Fast mode consumes the included allowance at 2.5× and Astra Ultrafast at 8× (credits: 2× / 6×) · Pro 500 additionally returns $500/mo in OpenAI API credits on a linked API organization — a separate API budget (prorated first grant, expires at the end of the billing cycle, no rollover) that excludes Codex usage through API keys and does not raise ChatGPT or Codex usage limits · Third-party apps use the plan through Sign in with ChatGPT (per-app weekly cap; no API key)
GPT-6 AstraGPT-6.1 SolGPT-6 SolGPT-6 LunaGPT-5.6 Sol ProGPT-5.6 SolGPT-5.6 TerraGPT-5.6 Luna
Ceiling
Lowest included usage of the three Pro tiers
Reachability
—
Pricing notes
Pro plans have no five-hour limit; OpenAI publishes local-message estimates for Plus and Standard Business only · local messages and cloud chats share the allowance, and weekly limits may also apply · Pro 100 additionally returns $100/mo in OpenAI API credits on a linked API organization — a separate API budget (prorated first grant, expires at the end of the billing cycle, no rollover) that excludes Codex usage through API keys and does not raise ChatGPT or Codex usage limits · Third-party apps use the plan through Sign in with ChatGPT (per-app weekly cap; no API key)
Cost rule
≈coarse estimate · platform.openai.com/docs/pricing — gpt-6-sol row ($2/$10, cached input $0.20, cache write $2.50; verified 2026-09-23); Codex availability: developers.openai.com/codex/pricing lists GPT-6 Sol at 15-150 local messages per 5h on its Plus and Standard Business table and states Pro plans currently have no five-hour limit; it publishes no Pro message estimate (read 2026-10-06)
GPT-6 AstraGPT-6.1 SolGPT-6 SolGPT-6 LunaGPT-5.6 SolGPT-5.6 TerraGPT-5.6 Luna
Ceiling
Unlimited* messages (reasonable use)
Reachability
—
Pricing notes
Codex usage estimates (Plus column, local messages per 5h): GPT-6 Luna 350–3,000 = highest · GPT-6.1 Sol 15–160 · GPT-6 Sol 15–150 · Astra 5–45 · weekly limits may also apply · Luna Reserve = extra Luna-only usage once the regular weekly cap is exhausted (selected Plus accounts, own limit) · Third-party apps use the plan through Sign in with ChatGPT (per-app weekly cap; no API key)
Cost rule
≈coarse estimate · platform.openai.com/docs/pricing — gpt-6-sol row ($2/$10, cached input $0.20, cache write $2.50; verified 2026-09-23); Codex availability: developers.openai.com/codex/pricing lists GPT-6 Sol at 15-150 local messages per 5h on its Plus and Standard Business table and states Pro plans currently have no five-hour limit; it publishes no Pro message estimate (Plus column re-read 2026-10-06; the calculator stores the low end of the range)
OzorePro$70 in credits/mo · + Smart Compression, Analytics Pro, Privacy Pack, 10 extra API keys, Custom Router, Fusion
$70 in credits/mo · + Smart Compression, Analytics Pro, Privacy Pack, 10 extra API keys, Custom Router, Fusion
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Includes Custom Router, so you can pin a lineup and Ozore routes only among your picks. Cached-read pricing is not published, so the calculator estimates it at 10% of the input rate.
Cost rule
~proxy estimate · floor: cached reads bill at the input rate · ozore.com/api/credits/pricing (live 2026-09-23, billed row openai/gpt-6-sol $1.30/$6.50 per 1M = 35% off list $2.00/$10.00; the row sits in the endpoint’s premium tier, served by both plans); cached-read not published → cached tokens billed at the input rate
OzoreBasic$20 in credits/mo · spend on any model at discounted rates · cancel anytime
$20 in credits/mo · spend on any model at discounted rates · cancel anytime
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Requests are auto-routed — you do not pick the model, so a per-model figure is an upper bound on this tier; Direct Pin and Custom Router are Pro features. Cached-read pricing is not published, so the calculator estimates it at 10% of the input rate.
Cost rule
~proxy estimate · floor: cached reads bill at the input rate · ozore.com/api/credits/pricing (live 2026-09-23, billed row openai/gpt-6-sol $1.30/$6.50 per 1M = 35% off list $2.00/$10.00; the row sits in the endpoint’s premium tier, served by both plans); cached-read not published → cached tokens billed at the input rate
Bills $3.99/mo after the free month · cancel anytime · entry tier (Flash-tier models) · one free trial month per person, duplicate trial accounts will be canceled · usage docs scope every plan to the same six-model always-on Flash floor (deepseek-v4.1-flash-1017, deepseek-v4.1-flash, glm-5.3-flash, mimo-v2.6-flash, deepseek-v4-flash-0731, deepseek-v4-flash) — the tier card names these five
Base shared usage pool (weekly refresh) · top-ups past the pool
Reachability
Cursor IDE / CLI / cloud agents only
Pricing notes
Only Grok models are available on this plan; Grok Bot is not included · The X Subscription ladder — sold by X Subscription Corp. (cursor.com/pricing) — bundles Cursor, Grok and X; SuperGrok plans are a separate purchase
Codex: GPT-6 Luna at Standard speed in the desktop app (subject to rollout) — Go does not get GPT-6 Sol · Third-party apps use the plan through Sign in with ChatGPT (per-app weekly cap; no API key)
List price $8/mo; the limited-time new-user price $3/mo is limited to accounts that have never purchased a Token Plan Personal Edition subscription (first subscription only) and is available until 23:59:59 on 2026-11-30 (UTC+8).
Cost rule
n/a
SyntheticGoWeekly credit quota, 2% regenerated every ~3.4h · no 5h or monthly caps · 1 usage reset/month · extra tokens at API pricing
Annual Pro plans bought before 2026-06-01 stay on legacy premium-request billing (premium requests, not AI credits) with a frozen model set — no new models or features until the term ends, then the subscriber drops to Copilot Free; the $15/mo AI-credit ceiling describes the monthly plan (docs.github.com en/copilot/reference/copilot-billing/request-based-billing-legacy, read 2026-10-23)
Usage caps are per model $15–$100/mo (each model’s 5h window = 20% and weekly = 50% of its cap): V4.1 Flash $100, $60 for GLM-5.2/5.3-Flash, Kimi K2.7 Code, LongCat-2.0, MiMo V2.6 Flash, MiMo V2.5, MiniMax M3, Muse Spark 1.2/1.3, Qwen3.7 Plus, Hy4 and Hy3, $30 for Qwen3.8 Flash, Hy4 preview and Gemini 4 Argon, $20 for MiniMax M3.1 Flash, $15 for V4 Pro, Grok 4.7, GLM-5.3, Kimi K3.1 Pro, Kimi K3, Step 5 Preview, Qwen3.8 Max, MiMo V2.6/V2.5 Pro, GPT-6/5.6 Luna and Claude Haiku 5.5 · retention 0d (Grok 4.7, Luna 30d)
60M input tokens/week · 2x on GLM 5/5.1/5.2, Kimi K2.7 Code, DeepSeek V4 Pro/Pro 0813 · 100 images/day
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Not the best fit for coding agents — large prompts/context burn the weekly cap fast; PAYG safer for heavy coding. Web search is not included in PRO’s subscription coverage.
Mistral Large 4Mistral Medium 3.5Mistral Small 4Mistral Large 3CodestralMinistral 3 14BMinistral 3 8BMinistral 3 3B
Ceiling
Fair usage · up to 6× Free · 150 Flash answers/day
Reachability
Vibe web/CLI/IDE (own client) · plan usage shared with the API
Pricing notes
Mistral Large 4 (Public Preview, 2026-10-06) supersedes Mistral Medium 3.5 and Mistral Large 3: both are marked deprecated since 2026-10-31 and stay served through 2027-04-30.
Price confirmed on the provider product page 2026-10-15 (dev.meta.ai/products/muse-code) · High Usage: 5× Everyday usage, more prompts with the latest Muse models
List price $16/mo; the limited-time new-user price $5/mo is limited to accounts that have never purchased a Token Plan Personal Edition subscription (first subscription only) and is available until 23:59:59 on 2026-11-30 (UTC+8).
Cost rule
n/a
KimiPlus1× agent credits · one pool shared with Kimi Code · rolling 5-hour rate window
$20/mo compute budget · no 5h/week windows · one agent at a time
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Dollar budget spent at provider list rates; no overage charges — requests stop at the budget. Plans carry no fee: every dollar of the price becomes compute, and top-ups cost usage + 5% (Economy and up).
Claude Fable 5.1Claude Fable 5Claude Sonnet 5Claude Opus 5.5
Ceiling
≥5× Free · 5h rolling window + weekly cap
Reachability
—
Pricing notes
Claude Fable 5 / Fable 5.1 are usage-credits-only on Pro (billed at API rates on top of the plan) — Fable is included usage on Max / Ultra at 50% of the weekly limit; Anthropic’s Usage Policy (effective 2026-11-12) prohibits reselling, proxying or otherwise providing access to Claude through unauthorized means, including “services that route requests through consumer subscriptions or misrepresent the product or client being used” (anthropic.com/legal/aup, read 2026-10-21)
Basic models plus rotating limited-time premium models. Non-production use (personal dev / prototyping — PAYG is the production tier). Flow ≈ $0.03283 (floats live); per-model Flow burn and per-tier model lists are not published in text. Provider "Worth" claim: $30/mo equivalent API value (≈1.5× the fee), from its published equivalent-pay-as-you-go table.
4× Plus shared usage pool · third-party models at API list price
Reachability
Cursor IDE / CLI / cloud agents only
Pricing notes
Grok Bot included; third-party models are not part of the included usage — they run on top-up credits billed separately at API list price · The X Subscription ladder — sold by X Subscription Corp. (cursor.com/pricing) — bundles Cursor, Grok and X; SuperGrok plans are a separate purchase
Cost rule
n/a
SyntheticStandardWeekly credit quota, 2% regenerated every ~3.4h · no 5h or monthly caps · 1 usage reset/month · extra tokens at API pricing
$41/mo compute budget · no 5h/week windows · parallel agents
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Dollar budget spent at provider list rates; no overage charges — requests stop at the budget. Plans carry no fee: every dollar of the price becomes compute, and top-ups cost usage + 5% (Economy and up).
The $10 Go plan’s higher tier: identical rates, per-model monthly caps $60–$240 (each model’s 5h window = 20% and weekly = 50% of its cap; the headline figure is the top of that range): V4.1 Flash $200, $240 for Hy3, Hy4 and LongCat-2.0, $180 for GLM-5.2/5.3-Flash, Kimi K2.7 Code, MiniMax M3 and Qwen3.7 Plus, $120 for GLM-5.3, MiMo V2.6 Flash, MiMo V2.5, Muse Spark 1.2/1.3, Hy4 preview and Gemini 4 Argon, $90 for Qwen3.8 Flash, $80 for MiniMax M3.1 Flash, $60 for V4 Pro, Grok 4.7, Kimi K3.1 Pro, Kimi K3, Step 5 Preview, Qwen3.8 Max, MiMo V2.6/V2.5 Pro, GPT-6/5.6 Luna and Claude Haiku 5.5 · retention 0d (Grok 4.7, Luna 30d)
Cost rule
n/a
Neural WattStandard6.25 kWh/mo energy incl. · priority
Price confirmed on the provider product page 2026-10-15 (dev.meta.ai/products/muse-code) · Power Usage: 20× Everyday usage, early access to new features
$95/mo compute budget · no 5h/week windows · priority scheduling
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Dollar budget spent at provider list rates; no overage charges — requests stop at the budget. Plans carry no fee: every dollar of the price becomes compute, and top-ups cost usage + 5% (Economy and up).
Cost rule
n/a
SyntheticBigWeekly credit quota, 2% regenerated every ~3.4h · no 5h or monthly caps · 1 usage reset/month · extra tokens at API pricing · higher priority access
Kimi K3DeepSeek V4.1 FlashNemotron 3 Super 120BQwen3.8-27BGPT-OSS 120BGLM-4.7-FlashGLM-5.3GLM-5.3-Flash
Ceiling
Weekly credit quota, 2% regenerated every ~3.4h · no 5h or monthly caps · 1 usage reset/month · extra tokens at API pricing · higher priority access
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Credits regenerate over the week and are spent at subscription-discounted model rates; more tokens can be bought at usage-based prices. Fast mode (2× speed; each request draws 2× the rate limit) and Infinite mode (Beta — no rate-limit draw, runs slower, DeepSeek V4.1 Flash only) are enabled from the Big plan
Cost rule
n/a
KimiMax5× agent credits · one pool shared with Kimi Code · rolling 5-hour rate window · 1M context
Claude Fable 5.1Claude Fable 5Claude Opus 5.5Claude Opus 4.8
Ceiling
5× Pro every 5 hours and every week · Claude Fable 5 / Fable 5.1 capped at 50% of the weekly limit
Reachability
—
Pricing notes
Claude Fable 5 / Fable 5.1 are included on Max but draw only 50% of the weekly limit (claude.com/pricing “Models and usage” matrix, read 2026-10-06); the plan also grants $100/month in Claude Platform API credits for building on the API — a separate budget from the plan’s Claude Code usage (anthropic.com/pricing “Monthly API credits”, read 2026-10-14); Anthropic’s Usage Policy (effective 2026-11-12) prohibits reselling, proxying or otherwise providing access to Claude through unauthorized means, including “services that route requests through consumer subscriptions or misrepresent the product or client being used” (anthropic.com/legal/aup, read 2026-10-21)
Cost rule
n/a
CursorSuper15× Plus shared usage pool · priority model access at peak times
15× Plus shared usage pool · priority model access at peak times
Reachability
Cursor IDE / CLI / cloud agents only
Pricing notes
Grok Bot included; third-party models are not part of the included usage — they run on top-up credits billed separately at API list price · The X Subscription ladder — sold by X Subscription Corp. (cursor.com/pricing) — bundles Cursor, Grok and X; SuperGrok plans are a separate purchase
Cost rule
n/a
Neural WattPro13.33 kWh/mo energy incl. · highest priority
Basic + premium models, covering the mainstream flagships. Non-production use (personal dev / prototyping — PAYG is the production tier). Flow ≈ $0.03283 (floats live); per-model Flow burn and per-tier model lists are not published in text. Provider "Worth" claim: $180/mo equivalent API value (≈1.8× the fee), from its published equivalent-pay-as-you-go table.
Claude Fable 5.1Claude Fable 5Claude Opus 5.5Claude Opus 4.8
Ceiling
No 5-hour limit (verified accounts) · 10× Pro every week · Claude Fable 5 / Fable 5.1 capped at 50% of the weekly limit
Reachability
—
Pricing notes
Claude Fable 5 / Fable 5.1 are included on Ultra but draw only 50% of the weekly limit (claude.com/pricing “Models and usage” matrix, read 2026-10-06); the plan also grants $200/month in Claude Platform API credits for building on the API — a separate budget from the plan’s Claude Code usage (anthropic.com/pricing “Monthly API credits”, read 2026-10-14); Anthropic’s Usage Policy (effective 2026-11-12) prohibits reselling, proxying or otherwise providing access to Claude through unauthorized means, including “services that route requests through consumer subscriptions or misrepresent the product or client being used” (anthropic.com/legal/aup, read 2026-10-21)
Cost rule
n/a
CursorUltra50× Plus shared usage pool · earliest access to new products and models
50× Plus shared usage pool · earliest access to new products and models
Reachability
Cursor IDE / CLI / cloud agents only
Pricing notes
Grok Bot included; third-party models are not part of the included usage — they run on top-up credits billed separately at API list price · Legacy Cursor Ultra ($200/mo, no longer sold) keeps two monthly pools — Cursor Models and an included Other Models allowance whose size Cursor does not publish · The X Subscription ladder — sold by X Subscription Corp. (cursor.com/pricing) — bundles Cursor, Grok and X; SuperGrok plans are a separate purchase
Cost rule
n/a
Neural WattMax28.57 kWh/mo energy incl. · shed-last priority
Price confirmed on the provider product page 2026-10-15 (dev.meta.ai/products/muse-code) · Maximum Usage: 80× Everyday usage, everything from the Power Usage plan · at the limit, wait for it to refresh — no upgrade or top-up
All models (Max and Ultra currently cover nearly identical lists — the difference is quota, not model selection). Non-production use (personal dev / prototyping — PAYG is the production tier). Flow ≈ $0.03283 (floats live); per-model Flow burn and per-tier model lists are not published in text. Provider "Worth" claim: $480/mo equivalent API value (≈2.4× the fee), from its published equivalent-pay-as-you-go table.
$269/mo compute budget · no 5h/week windows · highest-priority scheduling
Reachability
Any agent · OpenAI-compatible key
Pricing notes
Dollar budget spent at provider list rates; no overage charges — requests stop at the budget. Plans carry no fee: every dollar of the price becomes compute, and top-ups cost usage + 5% (Economy and up).
Cost rule
n/a
No plans in the database list this model.
Some Subscribe buttons carry referral or affiliate codes — using them may give you a signup bonus or support this project at no extra cost to you.
Markers: ~ proxy estimate · ≈ coarse estimate · ≥ cache floor · ≤ cache ceiling — a plan that publishes no cached-read rate bills cached tokens at the input rate, so its Req/$1 is a floor (≥) and its cost is a ceiling (≤). Borrowed cap means the budget is a published API cap at list price, so Req / $1 is a value comparison rather than a measured allowance.
Best-priced plans
Best-priced plans for GPT-6 Sol
The top 5 plans by computed Req/$1 that include GPT-6 Sol, plus the lowest-cost plan. Computed at build from the same cost rules the calculator renders · ~ proxy estimate · ≈ coarse estimate · ≥ cache floor · ≤ cache ceiling
“n/a” = the provider publishes no price structure for GPT-6 Sol — the calculator renders the same honest n/a. ◆ best value = highest computed Req/$1.
What real work costs
What GPT-6 Sol costs on real coding tasks
Illustrative cost of common coding tasks on the best-priced plans above, scaled from each plan's live per-request price to a task-typical token volume. Task token assumptions are shared ~proxy estimates — real tasks vary, and they are never per-provider shapes.
Task
LLM Gateway DevPass Pro
LLM Gateway DevPass Max
GitHub Copilot Max
LLM Gateway DevPass Lite
GitHub Copilot Pro+
Ozore Basic
Quick lookup / one-liner8,000 in · 1,000 out
~1¢
~1¢
~1¢
~1¢
~1¢
≤~1¢
Review a 500-line PR60,000 in · 4,000 out
~2¢
~2¢
~2¢
~2¢
~2¢
≤~8¢
Fix a bug (agent loop)180,000 in · 12,000 out
~5¢
~5¢
~5¢
~5¢
~5¢
≤~25¢
Refactor a module320,000 in · 20,000 out
~9¢
~9¢
~9¢
~9¢
~9¢
≤~45¢
Full-repo agent run900,000 in · 45,000 out
~25¢
~25¢
~25¢
~25¢
~25¢
≤~$1.24
Per-task cost = the plan's effective 50,010-token request price scaled to each task's token volume at the same shape mix. ~ = proxy estimate — shared task-token assumptions, not a quote. “n/a” = GPT-6 Sol unpriced on that plan. A cost marked ≤ is a cache ceiling: the real cost is that number or lower.
Recent changes
What changed for plans serving GPT-6 Sol
The 3 newest entries in the change ledger for plans that include GPT-6 Sol — each one dated and linked to its record.
The best plan pricing GPT-6 Sol delivers ~153 Req/$1 at the model's typical request shape — on LLM Gateway DevPass Pro. Figures are proxy confidence: they are computed from published rates and usage caps, proxy estimates are marked with a ~, and borrowed-cap plans use a published API cap at list price, so their Req/$1 is a value comparison rather than a measured allowance.
What's the cheapest plan with GPT-6 Sol?
Ozore Basic at $10/mo. The cheapest plan is not always the best value — check Req/$1 above.
Which plans include GPT-6 Sol?
18 tracked plans across 6 providers price it: CommandCode (Max 10×, Max 20×, Pro and Provider), OpenAI (Plus, Pro 100, Pro 200 and Pro 500), LLM Gateway DevPass (Lite, Pro and Max), Nous Portal (Plus, Super and Ultra), GitHub Copilot (Pro+ and Max) and Ozore (Basic and Pro). The calculator compares every tracked monthly plan that prices it.
Where does GPT-6 Sol rank on the Agent Arena leaderboard?
#12 (Max config) on the Agent Arena agent leaderboard — 7% net improvement (board fetched 2026-11-09). Ranks are scraped from https://arena.ai/leaderboard/agent, never typed by hand.
What is GPT-6 Sol?
GPT-6 Sol (API gpt-6-sol) is OpenAI's GPT-6-generation text generation model for complex coding and agentic work (family: openai, kind: text), released 2026-09-22 as the mid-tier beneath the GPT-6 Astra flagship.
How much does GPT-6 Sol cost?
GPT-6 Sol lists at $2 per 1M input tokens and $10 per 1M output tokens on OpenAI’s standard short-context rate card, with cached input at $0.20 and cache writes at $2.50. Requests above 272K input tokens bill as long context at 2x input and 1.5x output.
How does GPT-6 Sol compare to GPT-6 Astra?
GPT-6 Astra stays OpenAI’s frontier tier at $10/$50, while GPT-6 Sol is the cheaper GPT-6 tier beneath it at $2/$10 — positioned for demanding agentic coding and workflow automation rather than the hardest reasoning work.
Is GPT-6 Sol the same model as GPT-5.6 Sol?
No. GPT-6 Sol (gpt-6-sol) is the newer GPT-6-generation tier released 2026-09-22 at $2/$10; GPT-5.6 Sol (gpt-5.6-sol) is the earlier GPT-5.6 flagship at a $4/$20 list rate. They are separate models with separate API identifiers.
Is GPT-6 Sol open-weight?
No — GPT-6 Sol is a proprietary OpenAI model, reachable only through provider endpoints, so you cannot self-host it or access the weights. Open-weight agent models we track include GLM-5.3 and Hy4 preview.
Can I use GPT-6 Sol with any agent (BYOK)?
Yes — the tracked routers listing GPT-6 Sol serve it over OpenAI-compatible endpoints with no CLI lock, so it works with OpenCode, pi, Cursor or any agent that supports custom model endpoints.
A plain weekly digest of every tracked price move, ceiling change and model-list update — pulled straight from the dated changelog ledger, with a one-line “worth switching?” note on each. Free. No spam, unsubscribe anytime.