Kimi K3 (Reasoning) vs Claude Fable 5.1 (high)
Results updated 3 November 2026 · In Kimi Code and Claude Code, so part of any gap can come from the harness · How we benchmark
Verdict
Claude Fable 5.1 (high) beats Kimi K3 (Reasoning) by 5 points, and is faster; Kimi K3 (Reasoning) is cheaper.
- Cost per task
- $0.32 vs $0.44
- Time per task
- 1.5 min (estimated) vs 1.1 min
- Tokens per attempt
- 345k vs 132k
- Image input
- Yes vs No data
Kimi K3 (Reasoning), Kimi Code scores higher on Earth-Bench V2; Claude Fable 5.1 (high), Claude Code scores higher on GeoBenchX and GS-QA2. Kimi K3 (Reasoning), Kimi Code is cheaper ($0.32 against $0.44 per task), but Claude Fable 5.1 (high), Claude Code is faster (1.1 min against 1.5 min).
Where they differ most
By kind of work, share of tasks passed, 0 to 100 · Higher is better
| Kind of work | Kimi K3 (Reasoning) | Claude Fable 5.1 (high) |
|---|---|---|
| Remote sensing and raster analysis | 82 | 87 |
| GIS analysis | 84 | 88 |
| Change detection | 68 | 64 |
| Climate and time series | 80 | 76 |
Where the tasks are set: each kind of work links to its tasks and scoresCharts for each: Kimi K3 (Reasoning), Kimi Code · Claude Fable 5.1 (high), Claude Code.
Side by side
| Metric | Kimi K3 (Reasoning), Kimi Code | Claude Fable 5.1 (high), Claude Code |
|---|---|---|
| Geospatial Agent Index: Text only | 75 | 80 |
| Geospatial Agent Index: Text + vision | No data | 82 |
| Earth-Bench V2 score | 75 | 72 |
| GeoBenchX score | 65 | 80 |
| GS-QA2 score | 85 | 90 |
| Vision agents score | No data | 88 |
| GeoAgentBench score (Guided workflows; not in the Agent Index) | 96 | 100 |
| GeoAnalystBench score (Guided workflows; not in the Agent Index) | 100 | 95 |
| OGD4All score (standalone; in no index) | 97 | 97 |
| GeoNatureAgent score (standalone; in no index) | 100 | 100 |
| MultiGlobeQA score (standalone; in no index) | 77 | 79 |
| GeoAgentEval score (standalone; in no index) | 96 | 100 |
| ImpactMesh score (standalone; in no index) | No data | No data |
| Remote sensing and raster analysis score | 82 | 87 |
| Change detection score | 68 | 64 |
| GIS analysis score | 84 | 88 |
| Climate and time series score | 80 | 76 |
| Making maps score | 82 | 79 |
| Cost per task | $0.32 | $0.44 |
| Time per task (at 100 output tokens/s) | 1.5 min (estimated) | 1.1 min |
| Tokens per attempt | 345k | 132k |
| Turns per attempt | No data | 5.4 |
| Input price per 1M tokens | $3.00 | $10.00 |
| Output price per 1M tokens | $15.00 | $50.00 |
| Context window | 1,000,000 | 200,000 |
| Developer | Moonshot AI | Anthropic |
Frequently asked questions
Which scores higher on the Geospatial Agent Index, Kimi K3 (Reasoning), Kimi Code or Claude Fable 5.1 (high), Claude Code?
Claude Fable 5.1 (high), Claude Code, 80 to 75.
Which is cheaper per task, Kimi K3 (Reasoning), Kimi Code or Claude Fable 5.1 (high), Claude Code?
Kimi K3 (Reasoning), Kimi Code, $0.32 to $0.44.
Which writes fewer output tokens per task, Kimi K3 (Reasoning), Kimi Code or Claude Fable 5.1 (high), Claude Code?
Claude Fable 5.1 (high), Claude Code, 3k to 9k.
Which has the larger context window, Kimi K3 (Reasoning), Kimi Code or Claude Fable 5.1 (high), Claude Code?
Kimi K3 (Reasoning), Kimi Code, 1,000,000 tokens to 200,000 tokens.