Tiny Aya Earth vs Grok 4.20 (Non-Reasoning) vs Sarvam 105B
Grok 4.20 (Non-Reasoning) comes out ahead, 66 to 31 and 15 on our weighted score, though Sarvam 105B is 5.7× cheaper per token.
CohereTiny Aya Earth
15/100- ECI—
- Price—
- Context8K
- Our pick
xAIGrok 4.20 (Non-Reasoning)
66/100- ECI—
- Price$1.25 / $2.50
- Context1M
Sarvam AI
Sarvam 105B
31/100- ECI—
- Price$0.158 / $0.632
- Context131K
Grok 4.20 (Non-Reasoning) is our pick
Grok 4.20 (Non-Reasoning) is the better all-round choice, scoring 66/100 against Sarvam 105B (31) and Tiny Aya Earth (15). It leads on inputs & features and context window. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceSarvam 105BSarvam 105B $0.276 · Grok 4.20 (Non-Reasoning) $1.56 per 1M tokens (3:1 blend) · Tiny Aya Earth unpriced
- Longest contextGrok 4.20 (Non-Reasoning)Grok 4.20 (Non-Reasoning) 1,000,000 · Sarvam 105B 131,072 · Tiny Aya Earth 8,000 tokens
- Widest inputsGrok 4.20 (Non-Reasoning)Tiny Aya Earth: Text · Grok 4.20 (Non-Reasoning): Text, Images, PDFs · Sarvam 105B: Text
- Self-hostingTiny Aya Earth and Sarvam 105BPublishes downloadable weights (CC-BY-NC-4.0)
| Measure | Weight | Tiny Aya Earth | Grok 4.20 (Non-Reasoning) | Sarvam 105B |
|---|---|---|---|---|
| Inputs & features | 60% | 25 | 70 | 35 |
| Context window | 40% | 0 | 60 | 25 |
| Overall | 100% | 15/100 | 66/100 | 31/100 |
Left out because at least one model lacks the data: capability and price. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Tiny Aya EarthCohere | Grok 4.20 (Non-Reasoning)xAI | Sarvam 105BSarvam AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | — | $1.25 | $0.158 (best) |
| Output | — | $2.50 | $0.632 (best) |
| Cached input | — | $0.20 | $0.053 (best) |
| Blended (3:1) | — | $1.56 | $0.276 (best) |
| Long-context rate | — | Over 200K: $2.50 / $5.00 | Same rate |
| Price source | — | Official xAI API | Official Sarvam AI API |
| Limits | |||
| Context window | 8,000 tokens | 1,000,000 tokens (best) | 131,072 tokens |
| Max output | 8,000 tokens | 30,000 tokens | 131,072 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | No |
| PDFs | No | Yes | No |
| Audio | No | No | No |
| Video | No | No | No |
| Reasoning | No | No | Yeslow · medium · high |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | Yes | No |
| Availability | |||
| Weights | OpenCC-BY-NC-4.0 | Proprietary | Open |
| API model ID | tiny-aya-earth | grok-4.20-0309-non-reasoning | sarvam-105b |
| API providers | 1 | 13 (best) | 3 |
| Released | Feb 17, 2026 | Mar 9, 2026 | Feb 18, 2026 |
| Knowledge cutoff | — | — | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Tiny Aya Earth—
- Grok 4.20 (Non-Reasoning)$17.50
- Sarvam 105B$2.84
Which should you choose?
Which is better: Tiny Aya Earth, Grok 4.20 (Non-Reasoning) or Sarvam 105B?
Grok 4.20 (Non-Reasoning) is the better all-round choice, scoring 66/100 against Sarvam 105B (31) and Tiny Aya Earth (15). It leads on inputs & features and context window. The score weighs inputs & features 60%, context window 40%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Tiny Aya Earth, Grok 4.20 (Non-Reasoning) or Sarvam 105B?
Sarvam 105B is cheaper at $0.158 input / $0.632 output per million tokens (official Sarvam AI API price). Grok 4.20 (Non-Reasoning) costs $1.25 input / $2.50 output per million tokens (official xAI API price). At a typical mix of three input tokens to one output token, that is $0.276 per million tokens for Sarvam 105B versus $1.56 for Grok 4.20 (Non-Reasoning) (5.7× as much). Tiny Aya Earth has no published per-token price.
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Tiny Aya Earth has not been scored yet, Grok 4.20 (Non-Reasoning) has not been scored yet and Sarvam 105B has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Tiny Aya Earth, Grok 4.20 (Non-Reasoning) and Sarvam 105B yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Grok 4.20 (Non-Reasoning) has the largest context window at 1,000,000 tokens, against 131,072 for Sarvam 105B and 8,000 for Tiny Aya Earth. Maximum output per response: Tiny Aya Earth up to 8,000, Grok 4.20 (Non-Reasoning) up to 30,000, Sarvam 105B up to 131,072 tokens.
Which can read images, PDFs, audio or video?
Tiny Aya Earth accepts text; Grok 4.20 (Non-Reasoning) accepts text, images and PDFs; Sarvam 105B accepts text. Grok 4.20 (Non-Reasoning) handles the widest range of inputs.
Are any of these open source?
Tiny Aya Earth and Sarvam 105B publishes its weights (CC-BY-NC-4.0) and can be self-hosted; Grok 4.20 (Non-Reasoning) is proprietary.
Which is newer?
Grok 4.20 (Non-Reasoning) is the newest, released Mar 9, 2026. Sarvam 105B came out Feb 18, 2026; Tiny Aya Earth came out Feb 17, 2026.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Nov 2, 2026.
