OpenRouter 模型 ↗ · 2026-09-23
openai/gpt-6-sol
Model release OpenRouter 模型 Source language: Chinese
OpenRouter 于 2026-09-23 收录该模型(收录时间不等于模型正式发布日期);上下文:1,050,000 tokens;输入价:$2.0000/百万 tokens;输出价:$10.0000/百万 tokens;定位(来源描述):GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the…
Full text
OpenAI: GPT-6 Sol
openai/gpt-6-sol
Compare
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional work, agentic coding, business workflow automation, and computer use, and is particularly strong at long-horizon software engineering tasks in real codebases. It approaches Astra-level factual reliability at a much lower cost and shares Astra's clearer, more concise communication style in technical and coding conversations.
Modalities
In / out price
$2 / $10per 1M
Context
1.1M
Released
Sep 22, 2026
Compare
Providers
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick: Balanced (price + speed), Nitro (fastest), or Floor (cheapest).
Latency / throughput
$2.00$10.00$0.202.65s89 tok/s
100.00%
$2.00$10.00$0.201.83s65 tok/s
100.00%
$2.00$10.00$0.202.33s48 tok/s
100.00%
Not used in Standard routing:Why these endpoints are not used
Flex
$1.00$5.00$0.101.75s116 tok/s
100.00%
$2.20$11.00$0.224.21s65 tok/s
100.00%
$2.20$11.00$0.227.18s102 tok/s
100.00%
$2.20$11.00$0.22----
100.00%
Fast
$4.00$20.00$0.403.17s118 tok/s
100.00%
ProviderInput / Output /M
Input $2.00 / Output $10.00
89 tok/s100.00% uptime
Input $2.00 / Output $10.00
65 tok/s100.00% uptime
Input $2.00 / Output $10.00
48 tok/s100.00% uptime
Not used in Standard routing:Why these endpoints are not used
Input $1.00 / Output $5.00
Flex
116 tok/s100.00% uptime
Input $2.20 / Output $11.00
65 tok/s100.00% uptime
Input $2.20 / Output $11.00
102 tok/s100.00% uptime
Input $2.20 / Output $11.00
-- tok/s100.00% uptime
Input $4.00 / Output $20.00
Fast
118 tok/s100.00% uptime
Pricing
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Weighted average
Input
$0.5422/M tokens
Output
$10.13/M tokens
GPT-6 Sol — Price history
Price history
No effective price data for this selection
$0.513$10.0787.7%60.9%
$0.548$10.0386.5%37.1%
$1.224$10.0055.4%1.3%
Not used in Standard routing:Why these endpoints are not used
Flex
$0.767$5.0041.9%0.2%
$0.841$11.0069.2%0.3%
$1.942$11.0030.9%0.1%
Fast
$3.823$20.0025.5%0.1%
————
ProviderInput / Output /M
Token share
Effective in $0.513 / Effective out $10.07
Token share
60.9%
Effective in $0.548 / Effective out $10.03
Token share
37.1%
Effective in $1.224 / Effective out $10.00
Token share
1.3%
Not used in Standard routing:Why these endpoints are not used
Effective in $0.767 / Effective out $5.00
FlexToken share
0.2%
Effective in $0.841 / Effective out $11.00
Token share
0.3%
Effective in $1.942 / Effective out $11.00
Token share
0.1%
Effective in $3.823 / Effective out $20.00
FastToken share
0.1%
Effective in — / Effective out —
Token share —
Performance
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
Throughput
118tok/s
P50, best across providers
Latency
1.75s
P50, best provider
Reasoning effortLatency / throughput
AutoExacto benchmarks
GPQA DiamondTAU-BenchAmazon Bedrock92.4%70.0%
Uptime
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Uptime (3d)The model was reachable. Request routed to a provider.
100.00%
Availability (3d)The model returned inference from any provider. Errors and empty responses count against it.
99.10%
Availability over the last 3 days
Last 72 hours
Availability 99.10%
3 Days Ago2 Days AgoYesterdayNow
Availability over the last 24 hours
OpenRouter Availability
99.78%
Without Routing
98.74%
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and provider routing options.
Benchmarks
Scores on standardized evaluations and this model’s own outputs on shared benchmark prompts. Higher percentages are better — and each rank shows where this model lands among the models evaluated on that source.
Benchmark score summary for OpenAI: GPT-6 Sol (Artificial Analysis and Vals AI)
SourceBenchmarkScore
Artificial AnalysisGPT-6 Sol (Medium) HLE41.0%
Artificial AnalysisGPT-6 Sol (Medium) AA-LCR82.3%
Artificial AnalysisGPT-6 Sol (Medium) GDPval-AA42.5%
Artificial AnalysisGPT-6 Sol (Medium) CritPt24.6%
Artificial AnalysisGPT-6 Sol (Medium) SciCode53.8%
Artificial AnalysisGPT-6 Sol (Medium) Terminal-Bench 4.018.7%
Artificial AnalysisGPT-6 Sol (Low) HLE34.9%
Artificial AnalysisGPT-6 Sol (Low) AA-LCR79.3%
Artificial AnalysisGPT-6 Sol (Low) GDPval-AA35.0%
Artificial AnalysisGPT-6 Sol (Low) CritPt16.3%
Artificial AnalysisGPT-6 Sol (Low) SciCode50.2%
Artificial AnalysisGPT-6 Sol (Low) Terminal-Bench 4.09.1%
Artificial AnalysisGPT-6 Sol (Max) Intelligence Index44.2
Artificial AnalysisGPT-6 Sol (Max) HLE47.9%
Artificial AnalysisGPT-6 Sol (Max) AA-LCR83.7%
Artificial AnalysisGPT-6 Sol (Max) GDPval-AA50.7%
Artificial AnalysisGPT-6 Sol (Max) CritPt30.9%
Artificial AnalysisGPT-6 Sol (Max) SciCode57.6%
Artificial AnalysisGPT-6 Sol (Max) Terminal-Bench 4.043.9%
Artificial AnalysisGPT-6 Sol (Max) AA-Omniscience Accuracy51.1%
Artificial AnalysisGPT-6 Sol (Max) AA-Omniscience Non-Hallucination Rate39.2%
Artificial AnalysisGPT-6 Sol (Xhigh) HLE46.3%
Artificial AnalysisGPT-6 Sol (Xhigh) AA-LCR81.3%
Artificial AnalysisGPT-6 Sol (Xhigh) GDPval-AA48.3%
Artificial AnalysisGPT-6 Sol (Xhigh) CritPt28.0%
Artificial AnalysisGPT-6 Sol (Xhigh) SciCode55.1%
Artificial AnalysisGPT-6 Sol (Xhigh) Terminal-Bench 4.030.3%
Artificial AnalysisGPT-6 Sol (Non-reasoning) HLE18.4%
Artificial AnalysisGPT-6 Sol (Non-reasoning) AA-LCR64.0%
Artificial AnalysisGPT-6 Sol (Non-reasoning) GDPval-AA37.7%
Artificial AnalysisGPT-6 Sol (Non-reasoning) CritPt4.0%
Artificial AnalysisGPT-6 Sol (Non-reasoning) SciCode47.3%
Artificial AnalysisGPT-6 Sol (Non-reasoning) Terminal-Bench 4.013.1%
Artificial AnalysisGPT-6 Sol (High) HLE44.1%
Artificial AnalysisGPT-6 Sol (High) AA-LCR83.7%
Artificial AnalysisGPT-6 Sol (High) GDPval-AA45.1%
Artificial AnalysisGPT-6 Sol (High) CritPt25.4%
Artificial AnalysisGPT-6 Sol (High) SciCode54.9%
Artificial AnalysisGPT-6 Sol (High) Terminal-Bench 4.026.3%
Vals AIGPT-6 Sol BioMysteryBench74.8%
Vals AIGPT-6 Sol Code Migration57.2%
Vals AIGPT-6 Sol CyberBench v1.178.0%
Vals AIGPT-6 Sol EMB71.5%
Vals AIGPT-6 Sol Finance Agent (v2)49.0%
Vals AIGPT-6 Sol Financial Coaching Bench62.6%
Vals AIGPT-6 Sol Harvey's Legal Agent Benchmark v11.7%
Vals AIGPT-6 Sol IOI82.6%
Vals AIGPT-6 Sol Legal Research Bench28.8%
Vals AIGPT-6 Sol MedCode47.1%
Vals AIGPT-6 Sol MedScribe82.0%
Vals AIGPT-6 Sol MysteryMechanism30.2%
Vals AIGPT-6 Sol ProgramBench2.0%
Vals AIGPT-6 Sol ProofBench v1.183.0%
Vals AIGPT-6 Sol Public Benefits Bench v1.156.6%
Vals AIGPT-6 Sol Vals RSI Index28.1%
Vals AIGPT-6 Sol SAGE v144.8%
Vals AIGPT-6 Sol Vals Science Index45.0%
Vals AIGPT-6 Sol SRE Bench30.5%
Vals AIGPT-6 Sol Tax Agent Bench15.0%
Vals AIGPT-6 Sol Terminal-Bench 4.044.4%
Vals AIGPT-6 Sol Terminal-Bench-Science30.0%
Vals AIGPT-6 Sol Vals Index57.5%
Vals AIGPT-6 Sol Vibe Code Bench v1.187.8%
Apps
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
1.
Codex
A coding agent that helps you build and ship with AI
137Btokens
2.
Claude Code
Claude Code is Anthropic's agentic coding tool that reads your entire codebase, plans and executes changes across files, runs tests, and iterates on failures, all from natural language prompts.
64.7Btokens
3.
Hermes Agent
Hermes Agent is an open-source, self-improving AI agent by Nous Research that runs persistently with memory across sessions, and builds reusable skills from experience. It comes with 40+ built-in tools, including web search, browser automation, and vision, plus scheduled automations and subagents.
44.9Btokens
4.
minitap
new
33.6Btokens
5.
pi
There are many coding agents, but this one is yours.
32Btokens
Rest of OpenRouter
1.69B
Top 5 Apps
464M
Activity
Token volume and request traffic to this model over time.
API
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
OpenAI: GPT Sol versions
Available through EU in-region routing
OpenAI: GPT-6 Sol is available through EU in-region routing. Requests sent to eu.openrouter.ai are decrypted in the EU and routed only to shared provider endpoints in Europe (the EU/EEA, the UK, and Switzerland).
Providers in EuropeAzurePrices in Europe
Azure: $2.20/M tokens input, $11/M tokens output
Base URLhttps://eu.openrouter.ai/api/v1PlanRequires Business or Enterprise. Self-serve on Business: no contract, no monthly minimum. OpenRouter Business for EU in-region routing
Fails closed: if no shared provider endpoint in Europe is available, the request errors instead of leaving the region. Requests served by your own BYOK key go to your deployment’s region, which you control. In-region routing docs
Documentation
Guides, cookbooks, blog posts, and tools for building with OpenAI: GPT-6 Sol on OpenRouter.
GPT-6 Migration GuideCookbook
Frequently asked questions
GPT-6 Sol is the cost-efficient high-end model in OpenAI's GPT-6 series, positioned below the flagship GPT-6 Astra and above the fast GPT-6 Luna tier. It is suited for demanding professional work, agentic coding, business workflow automation, and computer use, and is particularly strong at long-horizon software engineering tasks in real codebases.
The GPT-6 Sol documentation on OpenRouter includes the GPT-6 Migration Guide cookbook. The quickstart covers API keys and your first request, and the API reference lists every parameter.
GPT-6 Sol costs $2.00/M input tokens and $10.00/M output tokens, with separate rates for Cache Read at $0.20/M tokens and Cache Write at $2.50/M tokens.
GPT-6 Sol has a 1,050,000 token context window. It supports up to 128,000 completion tokens.
Yes. GPT-6 Sol accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
Yes. OpenAI caches prompts of 1024 tokens or more automatically, and GPT-6 Sol bills a cache write on each cached prompt whether or not the cached prefix is read again. To turn caching off for a request, set prompt_cache_options.mode to "explicit" at the request root and do not mark any content block with prompt_cache_breakpoint. Explicit mode disables OpenAI-managed breakpoints, so a request with none of your own has no cacheable prefix and is billed at the plain input rate. See the prompt caching guide.
GPT-6 Sol accepts files such as PDFs, images, and text as input and returns text.
GPT-6 Sol is served by 3 providers on OpenRouter: OpenAI, Amazon Bedrock, and Azure. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
GPT-6 Sol was released on September 22, 2026.