Qwen3-Coder-NextvsGrok 4.20 Beta
Qwen3-Coder-Next | Grok 4.20 Beta | |
|---|---|---|
| Specifications | ||
ParametersA rough measure of how big the model is. More parameters usually means more capable and more expensive to run, though it is a poor guide on its own — a smaller, newer model often beats a larger, older one. | 80B | — |
Context windowHow much text the model can hold in mind at once — your question, any documents you attach, the conversation so far, and its own reply. Go past it and the earliest part falls out of view. | 256k | — |
| API pricingUSD per 1M tokens · lower wins · base tier | ||
Input priceWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it. | $0.30 | $1.25 |
Output priceWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one. | $1.50 | $2.50 |
Cached input priceA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again. | — | $0.20 |
Cheapest inputLowest input rate across third-party providers, excluding the lab itself. The cheapest endpoint may run a quantised build or a shorter context — see "Available from" on the model page. | $0.12Parasail | — |
Cheapest outputLowest output rate across third-party providers, excluding the lab itself. May come from a different provider than the cheapest input. | $0.80Parasail | — |
| Benchmarks | ||
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 70.6% | 76.7% |
| BenchmarksPublished by one model only | ||
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better. | — | 56% |
SWE-Bench ProAgentic coding — Can the AI fix real bugs in real software? It's handed actual problems from open-source projects and has to write code that genuinely solves them. Higher is better. | 44.3% | — |
SWE-Bench MultilingualMultilingual coding — Like SWE-Bench, but the coding problems span many programming languages, not just one. Tests how broadly the AI can code. Higher is better. | 62.8% | — |
Terminal-Bench 2.0Agentic terminal coding — Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? (Version 2.0 of the test.) Higher is better. | 36.2% | — |
ARC-AGI-2Abstract reasoning — Puzzle-style tests of abstract reasoning and pattern-finding — the kind of thing people find easy but AIs often struggle with. Higher is better. | — | 53.3% |
| Overview | ||
| Company | Qwen | SpaceXAI |
| Release date | Feb 3 2026 | Feb 17 2026 |
| Access | Open Weight | Proprietary |
Other comparisons
Qwen3-Coder-NextvsClaude Fable 5.1Grok 4.20 BetavsClaude Fable 5.1Qwen3-Coder-NextvsGPT-6 AstraGrok 4.20 BetavsGPT-6 AstraQwen3-Coder-NextvsGemini 3.8 FlashGrok 4.20 BetavsGemini 3.8 FlashQwen3-Coder-NextvsMuse Spark 1.3Grok 4.20 BetavsMuse Spark 1.3Qwen3-Coder-NextvsDeepSeek-V4-Pro-0813Grok 4.20 BetavsDeepSeek-V4-Pro-0813Qwen3-Coder-NextvsMistral Medium 3.5Grok 4.20 BetavsMistral Medium 3.5
Frequently asked questions
Grok 4.20 Beta leads Qwen3-Coder-Next on 1 of the 1 benchmark they both report (SWE-Bench Verified). Qwen3-Coder-Next is cheaper on both input and output: $0.30 vs $1.25 per million input tokens, and $1.50 vs $2.50 per million output tokens. Figures are base-tier rates. Qwen3-Coder-Next shipped 14 days before Grok 4.20 Beta, so benchmark comparisons should account for the intervening progress.
Qwen3-Coder-Next is open weight, while Grok 4.20 Beta is proprietary.
On SWE-Bench Verified, Grok 4.20 Beta leads at 76.7% vs Qwen3-Coder-Next at 70.6%.