Updated November 6, 2026. Rank says GPT-6 Astra is ahead. Price, access, and your workload can each overturn that. Public scores include evidence status and uncertainty. This is a same-family comparison, so migration details appear when the source data supports them.
GPT-6 Astra has the higher public score estimate, 85.8 versus 78.09, but the 90% score intervals overlap. Treat that as a lead, not a settled winner. 24 results are shared. Category rows resting on Estimated evidence or different benchmark sets are marked directional and do not name a winner.
Recommendations appear only when a shared evidence basis or an explicit operating constraint supports the call. Secondary and unsupported use cases stay disclosed below the initial list.
Code generation, repair, and software-engineering tasks
GPT-6 Astra
GPT-6 Astra has the higher public coding point estimate, 74.8 to 65.2, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
Tool use, computer use, and multi-step task completion
GPT-6 Astra
GPT-6 Astra has the higher public agentic point estimate, 73.3 to 57.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
1K fresh input + 500 output tokens
GPT-6 Sol
GPT-6 Sol has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
GPT-6 Sol
GPT-6 Sol has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
50K fresh input + 3K output tokens
GPT-6 Sol
GPT-6 Sol has the lower estimated token cost for this stated workload. Costs use the listed standard API rates.
Prompts that approach the documented context limit
No clear pick
The documented context windows are equal.
Choose a job from the LLM Selector's task catalog to see the category row it rests on, under the same basis rules as the table below. A directional row stays directional; choosing a job never creates a winner.
The same task catalog as the LLM Selector. Each job names the evidence surface it rests on; nothing here adds a new score.
Like-for-like · BenchAlign v5.8
GPT-6 Astra has the higher coding point estimate. Conditional score ranges do not establish rank confidence.
Coding scores combine specific tasks and setups. Match the editor, harness, and effort to your workflow.
Shared results can support a head-to-head reading. Results present for only one model describe coverage, not superiority.
1 category rests on Estimated evidence or different evidence sets. Those rows remain visible for coverage context but do not name a winner.
Only shared public evidence is shown. Sparse evidence stays a ruled list rather than being closed into a radar shape.
Too few matched category axes support a radar. The ruled list below shows only shared benchmark results; positions use each benchmark’s normalized display scale when available.
ARC-AGI-3Reasoning
Normalized gap 58.1OSWorld 2.0Agentic
Normalized gap 12.1AutomationBenchAgentic
Normalized gap 8.2FrontierMath v2 (Tier 4)Math
Normalized gap 7.6ARC-AGI-2Reasoning
Normalized gap 5.4Each row shows the public-lane category score for both models: the BenchAlign v5.8 lane for agentic, coding, and knowledge, and the weighted public rows elsewhere, with the same rank each model page reports. A row is like-for-like only when both scores rest on Supported evidence or the same weighted set. Directional and not-comparable rows remain visible, but they never receive a winner.
| Category | GPT-6 Astra | GPT-6 Sol | Basis | Reading |
|---|---|---|---|---|
| Agentic | 73.3Supported · #8/131 | 57.6Supported · #35/131 | Like-for-likeBenchAlign v5.8 lane · 16 vs 8 public rows | GPT-6 Astra leads · intervals overlap |
| Coding | 74.8Supported · #5/157 | 65.2Supported · #13/157 | Like-for-likeBenchAlign v5.8 lane · 8 vs 2 public rows | GPT-6 Astra leads · intervals overlap |
| Reasoning | 91.6#1/25 | 71.7#12/25 | Like-for-likeProvisional lane · 2 vs 2 weighted rows | GPT-6 Astra leads |
| Knowledge | 85.5Supported · #4/178 | 80.0Estimated · #9/178 | Directional onlyBenchAlign v5.8 lane · 7 vs 5 public rows | Directional only |
| Multimodal | 88.0Unranked · 3 rankable rows | 86.6Unranked · 1 rankable row | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Multilingual | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Instruction following | Not ranked | Not ranked | Not comparableProvisional lane · 0 vs 0 weighted rows | Not comparable |
| Math | 93.8Unranked · 2 rankable rows | 93.8Unranked · 2 rankable rows | Not comparableProvisional lane · 2 vs 2 weighted rows | Not comparable |
Ranks count the models scored in each category’s lane, so the agentic, coding, and knowledge denominators (BenchAlign v5.8) differ from the provisional-lane categories. Unranked scores sit on the provisional lane’s scale but never name a winner. Like-for-like rows within 0.5 points read as a practical tie.
Three fixed token mixes turn per-token rates into comparable decisions. Each scenario states context fit and whether cached input had to fall back to the published list-input rate.
1K fresh input + 500 output tokens
GPT-6 Sol has the lower modeled cost
Costs use the listed standard API rates.
50K fresh input + 3K output tokens
GPT-6 Sol has the lower modeled cost
Costs use the listed standard API rates.
200K cached + 20K fresh input + 10K output tokens
GPT-6 Sol has the lower modeled cost
Costs use the listed standard API rates.
Sourced differences are shown directly. Missing facts stay explicit instead of being inferred from a model name or family.
Maximum documented context; output-token limits may be lower.
GPT-6 Astra
GPT-6 Sol
GPT-6 Astra
gpt-6-astra
OpenAI GPT-6 Astra model documentationGPT-6 Sol
gpt-6-sol
OpenAI GPT-6 Sol model documentationA missing cached-input rate falls back to the listed input rate only in the stated workload estimate.
GPT-6 Astra
$1 per 1M cached input tokens
OpenAI GPT-6 Astra model documentationGPT-6 Sol
$0.2 per 1M cached input tokens
OpenAI GPT-6 Sol model documentationGPT-6 Astra
text, image
OpenAI GPT-6 Astra model documentationGPT-6 Sol
text, image
OpenAI GPT-6 Sol model documentationGPT-6 Astra
GPT-6 Sol
GPT-6 Astra
Limited Availability · OpenAI Responses API, OpenAI Chat Completions API, ChatGPT Plus, Pro, Business, and Enterprise, Microsoft Azure, AWS Bedrock
OpenAI GPT-6 Astra model documentationGPT-6 Sol
Generally Available · OpenAI Responses API, OpenAI Chat Completions API, ChatGPT Work, Codex
OpenAI GPT-6 Sol and Luna launchGPT-6 Astra
Reasoning
GPT-6 Sol
Reasoning
GPT-6 Astra
Proprietary
GPT-6 Sol
Proprietary
GPT-6 Astra
Proprietary
GPT-6 Sol
Proprietary
GPT-6 Astra
2026-09-03
GPT-6 Sol
2026-09-22
GPT-6 Astra has the higher public score estimate, 85.8 versus 78.09, but the 90% score intervals overlap. The higher estimate is not a decisive winner because the uncertainty ranges overlap.
GPT-6 Astra has the higher public coding point estimate, 74.8 to 65.2, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
GPT-6 Astra has the higher public agentic tasks point estimate, 73.3 to 57.6, with Supported evidence for both models. The conditional ranges do not establish rank confidence.
For the stated presets, chat costs $0.035 on GPT-6 Astra and $0.007 on GPT-6 Sol; repository review costs $0.65 and $0.13; the cache-heavy agent loop costs $0.9 and $0.18. Costs use the listed standard API rates.
Both models list the same context window, 1.05M.
The full public result ledger is available for audit without forcing a wide desktop table onto a phone.
LifeVal
Not directly comparable
BrowseComp
Not directly comparable
OSWorld 2.0
GPT-6 Astra leads this result
Terminal-Bench 4.0
Not directly comparable
Terminal-Bench-Science 0.1
Not directly comparable
ExploitGym
GPT-6 Astra leads this result
Agents' Last Exam
GPT-6 Astra leads this result
AutomationBench
GPT-6 Astra leads this result
HLE w/ tools
Not directly comparable
Terminal-Bench 2.1 (Vals)
Not directly comparable
ApprenticeBench
Not directly comparable
CWE-bench v1
Shared sourceGPT-6 Astra leads this result
Hermes Index
Shared sourceGPT-6 Astra leads this result
Hermes Bench
Shared sourceGPT-6 Astra leads this result
MazeBench
GPT-6 Astra leads this result
OSWorld 2.1
Not directly comparable
ScarfBench
Not directly comparable
DeepSWE
GPT-6 Astra leads this result
FrontierCode 1.1 Main
Not directly comparable
FrontierCode 1.1 Extended
Not directly comparable
FrontierSWE v2
Not directly comparable
PostTrainBench v1.1
Not directly comparable
Bug Hunt Bench
Not directly comparable
FrontierCode 2.0
Shared sourceGPT-6 Astra leads this result
BioBasicsBench
Not directly comparable
STEMBench
Not directly comparable
ARC-AGI-1
GPT-6 Astra leads this result
ARC-AGI-2
GPT-6 Astra leads this result
ARC-AGI-3
GPT-6 Astra leads this result
GeneBench-Pro
Not directly comparable
MRCR v2 256K-512K
Not directly comparable
MRCR v2 512K-1M
Not directly comparable
GraphWalks BFS 256K–1M
Not directly comparable
ARC-AGI-3 Provider Adapter
GPT-6 Astra leads this result
ExtractBench Pro
Not directly comparable
ExtractBench (LlamaIndex)
Shared sourceGPT-6 Sol leads this result
ParseBench
Shared sourceGPT-6 Astra leads this result
ScreenSpot Pro
Not directly comparable
BenchCAD Vision2Code (tools)
Not directly comparable
GPQA
Not directly comparable
GPQA-D
Not directly comparable
HealthBench (raw)
Shared sourceGPT-6 Astra leads this result
HealthBench (length-adjusted)
Shared sourceGPT-6 Astra leads this result
HealthBench Professional
Shared sourceGPT-6 Astra leads this result
HealthBench Professional (raw)
Shared sourceGPT-6 Astra leads this result
HealthBench Hard
Shared sourceGPT-6 Astra leads this result
Gray Swan IPI (15 attempts)
Shared sourceGPT-6 Astra leads this result
FrontierMath v2 (Tier 4)
GPT-6 Astra leads this result
FrontierMath v2 (Tiers 1-3)
Shared sourceGPT-6 Astra leads this result
One weekly email when material rank, price, or benchmark evidence changes make this matchup worth revisiting.
Read a sample issueJoin 6,100+ subscribers.
Last updated November 6, 2026
