GPT-6 Astra vs Inkling-Small: Specs & Benchmark Comparison

GPT-6 Astra is developed by OpenAI, while Inkling-Small comes from Thinking Machines Lab. Inkling-Small was released in July 2026, and GPT-6 Astra followed 2 months later in September 2026. Inkling-Small has a published size of about 276 billion parameters; OpenAI has not disclosed the parameter count of GPT-6 Astra.

The two models share 4 published benchmarks. GPT-6 Astra leads on 4 of them. The widest gaps are on ARC-AGI v2, where GPT-6 Astra scores 95.0% against 40.1%; ARC-AGI, where GPT-6 Astra scores 98.5% against 84.0%. Averaged across everything we track, GPT-6 Astra sits at 74.4% and Inkling-Small at 60.3%.

Inkling-Small is the cheaper API at $0.30 per million input tokens and $1.20 per million output tokens, roughly 33 times cheaper than GPT-6 Astra at $10 and $50. GPT-6 Astra takes the larger context window at 1M tokens, compared with 256K for Inkling-Small. GPT-6 Astra accepts text and images as input, while Inkling-Small accepts text, images, and audio. On tooling, only GPT-6 Astra supports function calling and only GPT-6 Astra offers structured output.

CharacteristicGPT-6 AstraInkling-Small
CompanyOpenAIThinking Machines Lab
Release DateSeptember 4, 2026July 30, 2026
Parameters—276B
MultimodalYesYes
Context (input)1.1M256K
Context (output)128K256K
Input Price / 1M$10.00$0.30
Output Price / 1M$50.00$1.20
Average Score74.4%60.3%
Benchmarks
ARC-AGI v295.0%40.1%
ARC-AGI98.5%84.0%
BrowseComp91.5%77.4%
GPQA Diamond96.0%89.5%

Visual Benchmark Comparison

GPT-6 Astra
Inkling-Small
ARC-AGI v20.9 vs 0.4
0.9
0.4
ARC-AGI1.0 vs 0.8
1.0
0.8
BrowseComp0.9 vs 0.8
0.9
0.8
GPQA Diamond1.0 vs 0.9
1.0
0.9

Verdict

GPT-6 Astra leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: GPT-6 Astra — 0.7, Inkling-Small — 0.6.

API Cost

Inkling-Small is 40.0x cheaper: input $0.30/1M vs $10.00/1M tokens.

Context Window

GPT-6 Astra supports a larger context: 1M vs 256K tokens.

Recency

GPT-6 Astra is newer: released 9/4/2026 vs 7/30/2026.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — GPT-6 Astra or Inkling-Small?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — GPT-6 Astra or Inkling-Small?
Inkling-Small is cheaper for input: $0.30 per 1M tokens vs $10.00.
Which has a larger context window — GPT-6 Astra or Inkling-Small?
GPT-6 Astra supports a larger context: 1,050,000 tokens vs 256,000.

The GPT-6 Astra and Inkling-Small comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the GPT-6 Astra or Inkling-Small page. See also the complete list of AI model comparisons.