grok-4-fast: benchmark scores, price and ranking
grok-4-fast from xAI ranks #267 of 743 LLMs on the LLMs Tiger aggregate benchmark score (49.5), based on 6 benchmark results captured up to 2026-10-13.
- Organisation
- xAI
- Type
- Frontier model
- Open weights
- No
- Aggregate rank
- #267 of 743 (score 49.5)
Relative to other models, grok-4-fast is strongest on Fiction.liveBench (#2 of 47) and weakest on WebDev Arena (#67 of 67).
Across the 6 benchmarks where at least ten models are scored, grok-4-fast is in the top 10% on 1 and in the top half on 5.
grok-4-fast benchmark results
| Benchmark | Score | Rank | Source | Captured |
|---|---|---|---|---|
| Fiction.liveBench | 94.4% | #2 of 47 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Epoch ECI | 144 | #30 of 204 | Epoch AI Benchmarking Hub | 2026-10-13 |
| DTBench | 82.7% | #43 of 146 | Epoch AI Benchmarking Hub | 2026-10-13 |
| Cybench | 30.0% | #6 of 16 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WeirdML | 42.9% | #49 of 109 | Epoch AI Benchmarking Hub | 2026-10-13 |
| WebDev Arena | 1,149 | #67 of 67 | Epoch AI Benchmarking Hub | 2026-10-13 |
About grok-4-fast
grok-4-fast is a frontier model from xAI. No API price is listed for it yet. We hold 6 benchmark results for it across 3 areas, and 1 is in the top tenth of its board.
Where grok-4-fast stands
On our aggregate score, grok-4-fast is #267 of 743 ranked models, around the middle of the table. Its score is 49.5; the scale is centred on 50, so it sits 0.5 points below the typical model. That beats 64% of the ranked field. Among the 26 ranked xAI models it is #13. The next one up is Grok-2 at 49.6. The next one down is Grok 3 mini at 49.4.
grok-4-fast, benchmark by benchmark
On Fiction.liveBench, grok-4-fast scores 94.4% and ranks #2 of 47. Only one model scores higher, led by grok-4-0709 at 94.4%, which is 0.0 points ahead. It covers long-context story comprehension (16k). The board median is 61.1%, so grok-4-fast is 33.3 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On Epoch ECI, grok-4-fast reaches 144. It ranks #30 of 204, ahead of 85% of the field, and 12 points behind Qwen 3.8 Max at 156. The results either side are Qwen3.7 Flash (145) and Gemini 2.5 Pro (Mar 2025) (144). It covers Epoch Capabilities Index: Epoch's own aggregate of ~40 benchmarks. The board median is 127, so grok-4-fast is 17 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
grok-4-fast manages 82.7% on DTBench, good for #43 of 146. Claude Opus 5.5 (max) leads at 98.9%, 16.3 points further up, and 71% of models sit below. The results either side are Gemma 4 31B IT (82.7%) and Qwen3.5 27B (82.4%). It covers dynamic tool-use benchmark. The board median is 68.8%, so grok-4-fast is 13.9 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Cybench puts grok-4-fast at 30.0%, #6 of 16. That beats 62% of the models tested, but 13.0 points short of grok-4-0709 (43.0%). The results either side are claude-sonnet-4-20250514 (35.0%) and claude-3-7-sonnet-20250219 (20.0%). It covers cybersecurity capture-the-flag tasks. The board median is 17.5%, so grok-4-fast is 12.5 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
On WeirdML, grok-4-fast reaches 42.9%. It ranks #49 of 109, ahead of 55% of the field, and 50.4 points behind GPT-6 Astra (max) at 93.3%. The results either side are Mercury 2 (43.2%) and Kimi K2 Thinking (42.8%). It covers unusual machine-learning tasks. The board median is 41.2%, so grok-4-fast is 1.7 points above it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
WebDev Arena is a weak spot. grok-4-fast scores 1,149, #67 of 67, behind 99% of the models measured. The leader, Claude Opus 5.5 (max), reaches 1,820. It covers building web apps, crowd-voted (Elo). The board median is 1,433, so grok-4-fast is 284 points below it. Source: Epoch AI Benchmarking Hub, captured 2026-10-13.
Strengths and weaknesses
General knowledge and chat: solid, ahead of 75% of the field on average across 4 benchmarks. Its best board here is Fiction.liveBench (#2 of 47) and its weakest is WeirdML (#49 of 109).
Reading these numbers
The 6 scores on this page come from 1 source: Epoch AI Benchmarking Hub (6). No official leaderboard supplied any of them, so treat them as third-party measurements. With 6 results the aggregate rank can move a lot when one new score arrives.
Models ranked nearby
- #264 Hunyuan-A13B-Instruct (49.5)
- #265 gpt-4o-2024-08-06 (49.5)
- #266 claude-opus-4-1-20250805_16K (49.5)
- #268 Grok 3 mini (49.4)
- #269 GPT-4 (0125) (49.4)
- #270 Gemini 3 Flash Preview (High) (49.4)
More from xAI
- Grok 4.6 (xhigh) (#34, 53.6)
- grok-4-0709 (#38, 53.2)
- Grok 4 (#39, 53.2)
- Grok 4.7 (xhigh) (#41, 53.1)
- Grok 4.5 (high) (#85, 52.0)
- Grok 4.20 (Reasoning) (#136, 51.2)
Compare it on the live leaderboard → · How the aggregate score works