Mistral Nemo Instruct 2407 vs GPT OSS 20B
GPT OSS 20B leads llmrun Score 96.4 to 65.4; both fit 24 GB at Q4_K_M; GPT OSS 20B is 2.3× faster on RTX 4090 (~190 vs ~81 tok/s).
Model A · Mistral AI
Mistral Nemo Instruct 240712.2B · 128K context · Apache 2.0 · Jul 2024
Model B · OpenAI
GPT OSS 20B20.9B · 4.2B active · 128K context · Apache 2.0 · Aug 2025
Choose Mistral Nemo Instruct 2407 if you want…
Lighter on memory
8.1 vs 12.9 GB at Q4_K_M
Choose GPT OSS 20B if you want…
Higher llmrun Score
96.4 vs 65.4
Ahead on the benchmarks
3 of 3 shared boards
2.3× faster on RTX 4090
~190 vs ~81 tok/s
Newer release
Aug 2025 vs Jul 2024
On your hardware
Each model at its default quant (Q4_K_M), a smaller one when it must. Estimates, ±20%.
8 GB
RTX 4060 · 8 GB tier
Mistral Nemo Instruct 2407
IQ2_M · ~37 tok/s
7.8 GB · up to 16K
GPT OSS 20B
IQ2_M · ~103 tok/s
7.6 GB · up to 32Kfaster
12 GB
RTX 3060 12GB · 12 GB tier
Mistral Nemo Instruct 2407
Q4_K_M · ~29 tok/s
11 GB · up to 16Kmore headroom
GPT OSS 20B
IQ4_XS · ~89 tok/s
11.8 GB · up to 16K
16 GB
RTX 5060 Ti 16GB · 16 GB tier
Mistral Nemo Instruct 2407
Q4_K_M · ~36 tok/s
11 GB · up to 32K
GPT OSS 20B
Q4_K_M · ~98 tok/s
13.1 GB · up to 128Kfaster
24 GBreference
RTX 4090 · 24 GB tier
Mistral Nemo Instruct 2407
Q4_K_M · ~81 tok/s
11 GB · up to 64K
GPT OSS 20B
Q4_K_M · ~190 tok/s
13.1 GB · up to 128Kfaster
32 GB
RTX 5090 · 32 GB tier
Mistral Nemo Instruct 2407
Q4_K_M · ~144 tok/s
11 GB · up to 64K
GPT OSS 20B
Q4_K_M · ~282 tok/s
13.1 GB · up to 128Kfaster
48 GB
RTX A6000 · 48 GB tier
Mistral Nemo Instruct 2407
Q4_K_M · ~62 tok/s
11 GB · up to 128K
GPT OSS 20B
Q4_K_M · ~154 tok/s
13.1 GB · up to 128Kfaster
Mac 32 GB
Mac Mini M4 · Macs
Mistral Nemo Instruct 2407
Q4_K_M · ~10 tok/s
11 GB · up to 64K
GPT OSS 20B
Q4_K_M · ~28 tok/s
13.1 GB · up to 128Kfaster
Mac 64 GB
Mac Studio M4 Max · Macs
Mistral Nemo Instruct 2407
Q4_K_M · ~47 tok/s
11 GB · up to 128K
GPT OSS 20B
Q4_K_M · ~86 tok/s
13.1 GB · up to 128Kfaster
Mac 128 GB
Mac Studio M4 Max · Macs
Mistral Nemo Instruct 2407
Q4_K_M · ~47 tok/s
11 GB · up to 128K
GPT OSS 20B
Q4_K_M · ~86 tok/s
13.1 GB · up to 128Kfaster
Reference GPU: RTX 4090. Tinted cell = the better pick on that machine (fits, more headroom, or at least 10% faster).
Overall
Mistral Nemo Instruct 2407
GPT OSS 20B
Coding
Mistral Nemo Instruct 2407
no data
GPT OSS 20B
Science
Mistral Nemo Instruct 2407
no data
GPT OSS 20B
Reasoning
Mistral Nemo Instruct 2407
no data
GPT OSS 20B
Mistral Nemo Instruct 2407 · #101 of 106 open models · #32 of 36 that fit 24 GB
GPT OSS 20B · #69 of 106 open models · #16 of 36 that fit 24 GB
Benchmarks
3 shared boards: Mistral Nemo Instruct 2407 leads 0, GPT OSS 20B leads 3 · Benchmarks imported Nov 2026
Mistral Nemo Instruct 2407
29.9%#84/100 open
IndependentEpoch AI · Jul 2024
GPT OSS 20B
60.8%#44/100 open
IndependentEpoch AI · Aug 2025
Mistral Nemo Instruct 2407
44.8%#87/146 open
IndependentTIGER-Lab
GPT OSS 20B
73.6%#33/146 open
IndependentTIGER-Lab
Mistral Nemo Instruct 2407
48.6%#65/70 open
Lab-reportedEpoch AI · Jul 2024
GPT OSS 20B
68.0%#36/70 open
Lab-reportedEpoch AI · Aug 2025
10 more boards (other shared and single-model boards)
Only Mistral Nemo Instruct 2407 (2)
- MATH Level 510.8%
- BALROG17.6%
Only GPT OSS 20B (8)
- LMArena Text1317.3
- AIME 2024/202565.3%
- Terminal-Bench3.4%
- SciCode38.9%
- WeirdML40.9%
- CritPt1.4%
- LMCA14.5%
- ALE-Bench566.0
VRAM by quantization
Memory to load the model and 2K context; below it, the total at 32K and 128K context. Lighter side in bold.
Q2_K
Mistral Nemo Instruct 2407
5.9 GB4.8 GB file
32K 12.2 GB · 128K 32.3 GB
GPT OSS 20B
9.3 GB11.5 GB file
32K 9.8 GB · 128K 11.5 GB
Q3_K_S
Mistral Nemo Instruct 2407
6.1 GB5.5 GB file
32K 12.4 GB · 128K 32.5 GB
GPT OSS 20B
9.5 GB11.5 GB file
32K 10.0 GB · 128K 11.7 GB
Q3_K_M
Mistral Nemo Instruct 2407
6.7 GB6.1 GB file
32K 13.0 GB · 128K 33.1 GB
GPT OSS 20B
10.6 GB11.5 GB file
32K 11.1 GB · 128K 12.8 GB
Q4_0
Mistral Nemo Instruct 2407
6.8 GB7.1 GB file
32K 13.1 GB · 128K 33.3 GB
GPT OSS 20B
10.8 GB11.5 GB file
32K 11.3 GB · 128K 13.0 GB
Q4_K_M
Mistral Nemo Instruct 2407
8.1 GB7.5 GB file
32K 14.4 GB · 128K 34.5 GB
GPT OSS 20B
12.9 GB11.6 GB file
32K 13.4 GB · 128K 15.1 GB
Q5_K_M
Mistral Nemo Instruct 2407
9.5 GB8.7 GB file
32K 15.7 GB · 128K 35.9 GB
GPT OSS 20B
15.3 GB11.7 GB file
32K 15.8 GB · 128K 17.5 GB
Q6_K
Mistral Nemo Instruct 2407
10.8 GB10.1 GB file
32K 17.1 GB · 128K 37.2 GB
GPT OSS 20B
17.6 GB12.0 GB file
32K 18.1 GB · 128K 19.8 GB
Q8_0
Mistral Nemo Instruct 2407
13.0 GB13.0 GB file
32K 19.3 GB · 128K 39.4 GB
GPT OSS 20B
21.3 GB~20.9 GB file
32K 21.8 GB · 128K 23.5 GB
19 more quants
IQ2_XXS
Mistral Nemo Instruct 2407
4.1 GB3.6 GB file
32K 10.4 GB · 128K 30.5 GB
GPT OSS 20B
6.1 GB~5.8 GB file
32K 6.6 GB · 128K 8.3 GB
IQ2_XS
Mistral Nemo Instruct 2407
4.4 GB3.9 GB file
32K 10.7 GB · 128K 30.8 GB
GPT OSS 20B
6.7 GB~6.3 GB file
32K 7.2 GB · 128K 8.9 GB
IQ2_S
Mistral Nemo Instruct 2407
4.6 GB4.1 GB file
32K 10.8 GB · 128K 31.0 GB
GPT OSS 20B
6.9 GB~6.5 GB file
32K 7.4 GB · 128K 9.1 GB
IQ2_M
Mistral Nemo Instruct 2407
4.9 GB4.4 GB file
32K 11.1 GB · 128K 31.3 GB
GPT OSS 20B
7.4 GB~7.1 GB file
32K 7.9 GB · 128K 9.6 GB
IQ3_XXS
Mistral Nemo Instruct 2407
5.5 GB4.9 GB file
32K 11.8 GB · 128K 31.9 GB
GPT OSS 20B
8.5 GB11.6 GB file
32K 9.0 GB · 128K 10.7 GB
Q2_K_S
Mistral Nemo Instruct 2407
not listed
GPT OSS 20B
8.7 GB12.1 GB file
32K 9.2 GB · 128K 10.9 GB
IQ3_XS
Mistral Nemo Instruct 2407
5.8 GB5.3 GB file
32K 12.1 GB · 128K 32.2 GB
GPT OSS 20B
9.0 GB11.6 GB file
32K 9.5 GB · 128K 11.2 GB
IQ3_S
Mistral Nemo Instruct 2407
5.9 GB5.6 GB file
32K 12.2 GB · 128K 32.3 GB
GPT OSS 20B
9.3 GB12.1 GB file
32K 9.8 GB · 128K 11.5 GB
IQ3_M
Mistral Nemo Instruct 2407
6.2 GB5.7 GB file
32K 12.5 GB · 128K 32.7 GB
GPT OSS 20B
9.8 GB11.6 GB file
32K 10.3 GB · 128K 12.0 GB
Q3_K_L
Mistral Nemo Instruct 2407
7.0 GB6.6 GB file
32K 13.3 GB · 128K 33.4 GB
GPT OSS 20B
11.1 GB11.5 GB file
32K 11.6 GB · 128K 13.3 GB
IQ4_XS
Mistral Nemo Instruct 2407
7.3 GB6.7 GB file
32K 13.6 GB · 128K 33.7 GB
GPT OSS 20B
11.6 GB11.6 GB file
32K 12.1 GB · 128K 13.8 GB
IQ4_NL
Mistral Nemo Instruct 2407
not listed
GPT OSS 20B
12.1 GB11.6 GB file
32K 12.6 GB · 128K 14.3 GB
Q4_1
Mistral Nemo Instruct 2407
not listed
GPT OSS 20B
12.1 GB11.6 GB file
32K 12.6 GB · 128K 14.3 GB
Q4_K_S
Mistral Nemo Instruct 2407
7.6 GB7.1 GB file
32K 13.9 GB · 128K 34.0 GB
GPT OSS 20B
12.1 GB11.6 GB file
32K 12.6 GB · 128K 14.3 GB
Q4_K_L
Mistral Nemo Instruct 2407
8.2 GB8.0 GB file
32K 14.5 GB · 128K 34.6 GB
GPT OSS 20B
13.2 GB11.9 GB file
32K 13.7 GB · 128K 15.4 GB
Q5_0
Mistral Nemo Instruct 2407
8.4 GB8.5 GB file
32K 14.7 GB · 128K 34.8 GB
GPT OSS 20B
not listed
Q5_K_S
Mistral Nemo Instruct 2407
9.1 GB8.5 GB file
32K 15.4 GB · 128K 35.6 GB
GPT OSS 20B
14.8 GB11.7 GB file
32K 15.3 GB · 128K 17.0 GB
Q5_K_L
Mistral Nemo Instruct 2407
9.6 GB9.1 GB file
32K 15.9 GB · 128K 36.0 GB
GPT OSS 20B
15.5 GB11.9 GB file
32K 16.0 GB · 128K 17.7 GB
BF16
Mistral Nemo Instruct 2407
25.2 GB~24.5 GB file
32K 31.5 GB · 128K 51.6 GB
GPT OSS 20B
42.2 GB~41.8 GB file
32K 42.7 GB · 128K 44.4 GB
Mistral Nemo Instruct 2407
Ollama · Q4_K_M
ollama run hf.co/bartowski/Mistral-Nemo-Instruct-2407-GGUF:Q4_K_MGPT OSS 20B
Ollama · Q4_K_M
ollama run hf.co/unsloth/gpt-oss-20b-GGUF:Q4_K_M| Spec | Mistral Nemo Instruct 2407 | GPT OSS 20B |
|---|---|---|
| Publisher | Mistral AI | OpenAI |
| Parameters | 12.2B | 20.9B |
| Active parameters | — | 4.2B per token |
| Architecture | MistralForCausalLM | GptOssForCausalLM |
| Context | 128K tokens | 128K tokens |
| Modalities | Chat, Function-calling | Chat, Function-calling |
| License | Apache 2.0 | Apache 2.0 |
| Released | Jul 2024 | Aug 2025 |
Frequently Asked Questions
Which is better, Mistral Nemo Instruct 2407 or GPT OSS 20B?
GPT OSS 20B leads llmrun Score 96.4 to 65.4; both fit 24 GB at Q4_K_M; GPT OSS 20B is 2.3× faster on RTX 4090 (~190 vs ~81 tok/s). Choose Mistral Nemo Instruct 2407 for lighter on memory; choose GPT OSS 20B for higher llmrun Score and ahead on the benchmarks.
Which needs less VRAM, Mistral Nemo Instruct 2407 or GPT OSS 20B?
At Q4_K_M, Mistral Nemo Instruct 2407 requires 8.1 GB while GPT OSS 20B needs 12.9 GB, making Mistral Nemo Instruct 2407 the easier model to run on your local setup.
Which is faster on RTX 4090?
Estimated decode speed on RTX 4090 is about 81 tok/s for Mistral Nemo Instruct 2407 (Q4_K_M) and 190 tok/s for GPT OSS 20B (Q4_K_M), so GPT OSS 20B is faster. Estimates, ±20%.
Can I run Mistral Nemo Instruct 2407 or GPT OSS 20B on a 16 GB GPU?
Mistral Nemo Instruct 2407 fits at Q4_K_M; GPT OSS 20B fits at Q4_K_M.
Which has a longer context window, Mistral Nemo Instruct 2407 or GPT OSS 20B?
Mistral Nemo Instruct 2407 supports 131,072 tokens and GPT OSS 20B supports 131,072 tokens.