Newest models
Models released since May 1, 2026. 27 of 338 catalog models match as of July 2026.
- 27
- Models
- $0.077
- Input from /1M
- 1.0M
- Max context
- Jun 30, 2026
- Newest release
27 models have shipped since May 5, 2026, and the pace hasn't let up: Claude Sonnet 5 and Nano Banana 2 Lite both landed June 30, 2026, the most recent entries in the list. Pricing spans the usual range once you exclude the placeholder-priced Fusion model — Ring-2.6-1T comes in at $0.075/$0.625 per million (input/output), while Claude Opus 4.7 (Fast) sits at $30/$150. Three models here are free: North Mini Code, Nemotron 3 Ultra (free), and Nemotron 3.5 Content Safety (free). Context windows have quietly crept up too — GLM 5.2 and MiniMax M3 both ship above 1M tokens, and Fugu Ultra and Claude Sonnet 5 hit the same mark. If you're evaluating what's new, look past the release date to what actually changed: context length, modality support, and where a model ranks on Design Arena.
Top picks
Claude Sonnet 5
The newest model in the catalog (created 2026-06-30), with a 1M-token context window and text+image+file input at $2/$10 per million tokens — a mid-tier price point for a frontier-dated release.
GLM 5.2
Ranks #1 on design_arena's 3D benchmark (Elo 1376), the highest score in this batch, at $0.93/$3.00 per million tokens with a 1,048,576-token context window.
Nemotron 3 Ultra (free)
Same 1M context window and design_arena 3D rank (40, Elo 1213) as its paid counterpart, but free — a way to test Nvidia's newest architecture at zero input/output cost.
All 27 models
Full catalog →Claude Sonnet 5Anthropic1Mctx 1M · inout
Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)Google66Kctx 66K · inout
Fugu UltraSakana AI1Mctx 1M · inout
Nano Banana Pro (Gemini 3 Pro Image)Google66Kctx 66K · inout
Nano Banana 2 (Gemini 3.1 Flash Image)Google131Kctx 131K · inout
North Mini Code (free)FreeCohere256KFreeFreectx 256K · inFreeoutFree
GLM 5.2Z.ai1.0Mctx 1.0M · inout
FusionOpenRouter1MVariableVariablectx 1M · inVariableoutVariable
Kimi K2.7 CodeMoonshot AI262Kctx 262K · inout
Claude Fable 5Anthropic1Mctx 1M · inout
Nex-N2-ProNex262Kctx 262K · inout
Nemotron 3 UltraNVIDIA1Mctx 1M · inout
Nemotron 3 Ultra (free)FreeNVIDIA1MFreeFreectx 1M · inFreeoutFree
Nemotron 3.5 Content Safety (free)FreeNVIDIA128KFreeFreectx 128K · inFreeoutFree
Qwen3.7 PlusQwen1Mctx 1M · inout
MiniMax M3MiniMax1.0Mctx 1.0M · inout
Step 3.7 FlashStepFun256Kctx 256K · inout
Claude Opus 4.8Anthropic1Mctx 1M · inout
Claude Opus 4.8 (Fast)Anthropic1Mctx 1M · inout
Qwen3.7 MaxQwen1Mctx 1M · inout
Grok Build 0.1xAI256Kctx 256K · inout
Gemini 3.5 FlashGoogle1.0Mctx 1.0M · inout
Claude Opus 4.7 (Fast)Anthropic1Mctx 1M · inout
Perceptron Mk1Perceptron AI33Kctx 33K · inout
Ring-2.6-1TInclusionAI262Kctx 262K · inout
Gemini 3.1 Flash LiteGoogle1.0Mctx 1.0M · inout
GPT Chat LatestOpenAI400Kctx 400K · inout
Prices per 1M tokens, flat 3% fee included. Hover or tap a price for the math.
How to choose
First, separate 'new' from 'better': GLM 5.2's Elo 1376 on design_arena 3D beats Claude Opus 4.8's Elo 1343 on androidnative, but they're not measuring the same task — match the benchmark to your actual use case. Second, context length varies more than you'd expect among same-week releases: Nemotron 3.5 Content Safety caps at 128,000 tokens while Fugu Ultra (released six days later) hits 1,000,000. Third, check modality before price — Gemini 3.5 Flash accepts text, image, file, audio, and video, while Ring-2.6-1T is text-only, and that constraint often matters more than the $1.5/$9.00 vs $0.075/$0.625 per-million gap.
Questions
- What's the newest model on OpenKey right now?
- Claude Sonnet 5, created 2026-06-30, tied with Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) for the most recent release date. Sonnet 5 runs $2/$10 per million tokens (input/output) with a 1M-token context window and text+image+file input.
- Are any of the newest models free to use?
- Yes — three of the 27 recent releases are free: North Mini Code, Nemotron 3 Ultra (free), and Nemotron 3.5 Content Safety (free). The Nemotron free tier matches its paid version's 1M context and benchmark rank, so it's worth testing before paying for the standard variant.
- How does OpenKey pricing work for these new models?
- OpenKey charges provider list price plus a flat 3% fee, no markup beyond that. For Claude Sonnet 5 at $2.00/$10.00 provider price, OpenKey pricing works out to $2.00 x 1.03 = $2.06 input and $10.00 x 1.03 = $10.30 output per million tokens.