Hermes 4 70B is 113.2× cheaper per million tokens (blended). Claude Fable 5 has the larger context window (1,000,000 vs 131,072 tokens, 7.6×). Hermes 4 70B is open-weights; the other is not.
Claude Fable 5 vs Hermes 4 70B
Live catalog fields. Best-in-row is highlighted.
Claude Fable 5
Anthropic · anthropic__claude-fable-5
Hermes 4 70Bopen
Nous Research · nousresearch__hermes-4-70b
Identity
| Field | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
|---|---|---|
| Provider | Anthropic | Nous Research |
| Slug | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
| Status | live | live |
| Open weights | No | Yes |
| License | — | — |
| HuggingFace | — | NousResearch/Hermes-4-70B |
Quality
| Field | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
|---|---|---|
| Quality index (0–100) | 89.4 | — |
Cost
| Field | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
|---|---|---|
| Input $ / 1M tokens | $10.00 | $0.13 |
| Cached input $ / 1M | $1.00 | — |
| Output $ / 1M tokens | $50.00 | $0.40 |
Context
| Field | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
|---|---|---|
| Max context tokens | 1,000,000 | 131,072 |
| Max output tokens | 128,000 | 117,964 |
Modalities
| Field | anthropic__claude-fable-5 | nousresearch__hermes-4-70b |
|---|---|---|
| Input | text, image, file | text |
| Output | text | text |