Model comparison
gpt-oss-20b vs Grok 4.1 Fast
Grok 4.1 Fast is the stronger model overall, scoring 40.9 to 32.1 on the Noometry Index. gpt-oss-20b costs 7.6× less per token, which makes it the better buy when Grok 4.1 Fast's lead doesn't matter for your workload.
Last verified . 19 shared benchmarks.
Summary
- They share 19 benchmarks with published results for both. gpt-oss-20b scores higher in 3 categories and Grok 4.1 Fast in 6 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Grok 4.1 Fast leads 36.1 to 9.2.
- The biggest single-benchmark swing is DTBench: 68% for gpt-oss-20b and 87.7% for Grok 4.1 Fast.
- gpt-oss-20b is cheaper at $0.018 / $0.09 per million input/output tokens, against $0.20 / $0.50 for Grok 4.1 Fast.
- gpt-oss-20b accepts more context: 131K tokens versus 128K.
- gpt-oss-20b has downloadable open weights; the other is API-only.
Side by side
| gpt-oss-20b | Grok 4.1 Fast | |
|---|---|---|
| Provider | OpenAI | xAI |
| Noometry Index | 32.1 | 40.9 |
| Released | 2025-08-05 | 2025-06-27 |
| Weights | Open | Proprietary |
| Context window | 131K | 128K |
| Max output | 16K | 30K |
| Input $ / M tokens | $0.018 | $0.20 |
| Output $ / M tokens | $0.09 | $0.50 |
| Results tracked | 34 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding gpt-oss-20b leads
gpt-oss-20b: 37.1 (#185), Grok 4.1 Fast: 33.8 (#236)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Coding | 1307 | 1411 |
| ALE-Bench | 566.05 | 394.93 |
| LMArena WebDev | — | 1241 |
| SciCode | 34.4% | — |
| WeirdML | 40.9% | — |
Agentic & Tool Use Grok 4.1 Fast leads
gpt-oss-20b: 9.2 (#146), Grok 4.1 Fast: 36.1 (#35)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| Terminal-Bench | 3.4% | — |
| Berkeley Function Calling Leaderboard | — | 69.6% |
| τ²-bench Banking | — | 13.1% |
| LMArena Search | — | 1171 |
| Vending-Bench 2 | — | 1,107 |
Reasoning Grok 4.1 Fast leads
gpt-oss-20b: 18.8 (#251), Grok 4.1 Fast: 42.3 (#45)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Hard Prompts | 1274 | 1407 |
| DTBench | 68% | 87.7% |
| SimpleBench | — | 56% |
| Kagi LLM Benchmark | 53.2% | — |
| NYT Connections (extended) | — | 87.4% |
| CritPt | 1.4% | — |
| Chess Puzzles | 4% | — |
| LMCA | 14.5% | — |
| Epoch Capabilities Index | 137.81 | — |
| ForecastBench | — | 61 |
Math gpt-oss-20b leads
gpt-oss-20b: 37.6 (#93), Grok 4.1 Fast: 30.9 (#201)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Math | 1317 | 1408 |
| MathArena Final-Answer Competitions | — | 60.9% |
| OTIS Mock AIME 2024-2025 | 65.3% | — |
| ProofBench | — | 4% |
| Omni-MATH | 56.5% | — |
Knowledge gpt-oss-20b leads
gpt-oss-20b: 34.6 (#186), Grok 4.1 Fast: 32.6 (#203)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Expert | 1258 | 1398 |
| GPQA Diamond | 60.8% | — |
| MMLU-Pro | 74% | — |
| Vectara Hallucination Rate | — | 17.8% |
| GPQA (HELM) | 59.4% | — |
Multimodal Not comparable
gpt-oss-20b: —, Grok 4.1 Fast: 36.3 (#71)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Vision | — | 1201 |
Multilingual Grok 4.1 Fast leads
gpt-oss-20b: 42.4 (#192), Grok 4.1 Fast: 51.1 (#110)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Non-English | 1269 | 1391 |
| LMArena Chinese | 1311 | 1441 |
| LMArena German | 1249 | 1404 |
| LMArena Japanese | 1245 | 1347 |
| LMArena Korean | 1239 | 1361 |
| LMArena Russian | 1279 | 1387 |
| LMArena Spanish | 1267 | 1411 |
| LMArena French | — | 1416 |
Instruction Following Grok 4.1 Fast leads
gpt-oss-20b: 62.0 (#235), Grok 4.1 Fast: 72.8 (#127)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Instruction Following | 1236 | 1376 |
| IFEval | 73.2% | — |
Long Context Grok 4.1 Fast leads
gpt-oss-20b: 37.9 (#203), Grok 4.1 Fast: 42.4 (#121)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Longer Query | 1251 | 1390 |
Writing & Preference Grok 4.1 Fast leads
gpt-oss-20b: 35.8 (#260), Grok 4.1 Fast: 57.4 (#126)
| Benchmark | gpt-oss-20b | Grok 4.1 Fast |
|---|---|---|
| LMArena Text | 1288 | 1408 |
| LMArena Creative Writing | 1201 | 1394 |
| EQ-Bench Creative Writing | 666 | 1327 |
| LMArena Multi-Turn | 1268 | 1389 |
| WildBench | 73.7% | — |
Frequently asked questions
Is gpt-oss-20b better than Grok 4.1 Fast?
Grok 4.1 Fast is the stronger model overall, scoring 40.9 to 32.1 on the Noometry Index. gpt-oss-20b costs 7.6× less per token, which makes it the better buy when Grok 4.1 Fast's lead doesn't matter for your workload.
Which is cheaper, gpt-oss-20b or Grok 4.1 Fast?
gpt-oss-20b is cheaper. It lists at $0.018 per million input tokens and $0.09 per million output tokens; Grok 4.1 Fast lists at $0.20 and $0.50.
Is gpt-oss-20b or Grok 4.1 Fast better for coding?
gpt-oss-20b scores higher on coding benchmarks: 37.1 versus 33.8 in the Noometry coding category.
Which has the bigger context window?
gpt-oss-20b does, with 131K tokens against 128K.
How many benchmarks do gpt-oss-20b and Grok 4.1 Fast share?
19 benchmarks have published results for both models. gpt-oss-20b has 34 scored results on Noometry and Grok 4.1 Fast has 32.